fix(ef): RG 476(f) -- the Creed is not said at a Requiem Mass

colitur said the Creed IS said on All Souls' Day and every other
Requiem-shaped day: RG 475(b)'s "in festis I classis" branch fired on
rank alone, with nothing checking RG 476(f), "Non dicitur symbolum ...
f) in Missis defunctorum". Allow-listed as L1 in
expected-divergences-lms.sexp on the argument that [creed] has no
notion of a Requiem Mass in its inputs and could not express the rule
without a signature change.

That argument does not hold. Colour.Black is used by exactly two
celebrations in the whole engine, verified by grepping every
occurrence in lib/ and data/: commemoration-of-all-souls
(data/ef/sanctoral.sexp) and Good Friday (temporal_ef.ml, already
false via the earlier RG 23(b) feria exclusion and has no Mass at all
in the 1955-restored Holy Week). So on every day this engine can
construct, colour = Black if and only if the Mass is a Requiem -- a
citable implication (RG 117 assigns black to Masses of the dead) on
this data, not a heuristic.

Rite_ef.Rubrics_ef.creed now reads observed.colour as RG 476(f)'s own
guard, positioned ahead of 475(b)'s rank branch. A new test,
test_colour_black_population_is_exactly_two, sweeps both the shipped
data and Temporal_ef's own code across a century and fails loudly the
day a third Colour.Black celebration is ever introduced, so the proxy
cannot silently rot.

L1 is closed in expected-divergences-lms.sexp, citation preserved.

Blast radius, measured (full 1583-9999 sweep, grouped by observed
slug): 8 416 days change from true to false, every one
commemoration-of-all-souls (missing from exactly 1 of the 8 417 domain
years, a boundary-year transfer effect at 1583/9999). A further 8 416
Colour.Black days are Good Friday, unaffected (already false before
this fix). colitur day/readings stay byte-identical to v0.10.1
(checked directly against a v0.10.1 worktree build, 9 domain-spanning
years); colitur rubrics changes on exactly the 8 416 days above.

colitur emit --format sexp's pretty-printed line count for 2027 moved
9010 -> 9011 as a side effect: Sexplib.Sexp.to_string_hum wraps by
column width, and All Souls' Day 2027's record grew one wrapped line
because "false" is one character longer than "true". Cosmetic only --
no field's data changed -- and pinned in test/cli.t with the
measurement recorded.

1 files changed, 91 insertions, 36 deletions

diff --git a/data/ef/expected-divergences-lms.sexp b/data/ef/expected-divergences-lms.sexp
index 96c9e50..e327419 100644
--- a/data/ef/expected-divergences-lms.sexp
+++ b/data/ef/expected-divergences-lms.sexp
@@ -15,41 +15,96 @@
 ; backs, the SAME convention the missalemeum file's own header states:
 ; not always "colitur".
 ;
-; L1 -- 2025-11-03 (All Souls' Day, transferred from 2 November, a Sunday
-; that year, to the next available day -- RG's own transfer of a
-; Commemoration when it falls on a Sunday). colitur (Rite_ef.Rubrics_ef.
-; creed) says the Creed IS said (true): observed.rank = Class1, so RG
-; 475(b)'s "in festis I classis" branch fires with nothing to stop it.
-; The Ordo says it is NOT said (false): "Mon 3 ALL SOULS' DAY I Cl B / No
-; Gl Sequence No Cr Pr of the Dead No Blessing" -- confirmed directly
-; against the extracted pdftotext text, not merely the fixture row.
+; L1 -- CLOSED, FIXED (whole-branch review finding 1, 2026-08-22): was
+; 2025-11-03 (All Souls' Day, transferred from 2 November, a Sunday that
+; year). colitur used to say the Creed IS said (true) on rank alone
+; (475(b), Class1) where the Ordo, correctly, says it is not: "Mon 3 ALL
+; SOULS' DAY I Cl B / No Gl Sequence No Cr Pr of the Dead No Blessing".
 ;
-; The Ordo is right and colitur is not, on the rubric itself: RG 476(f),
-; "Non dicitur symbolum ... f) in Missis defunctorum" -- the Creed is
-; never said at a Requiem Mass, and All Souls' Day is definitionally one
-; (colour black in colitur's own output on this very date). This is NOT a
-; new defect this task found and is choosing to leave open -- it is
-; rubrics_ef.ml's OWN documented scope limit, stated in that file's header
-; comment before this task existed: "476(f) (Requiem Masses, also not
-; modelled) ... genuinely inapplicable to this implementation -- a
-; documented scope limit, not a defect." [creed] has no notion of "this
-; Mass is a Requiem" anywhere in its own inputs (temporal/observed/date),
-; so it cannot express 476(f) without a signature change this task's own
-; constraints do not authorise (lib/ is read-only here except for this
-; allow-list). Recorded here as the SPECIFIC real-world day the scope
-; limit is finally witnessed on, rather than left as an abstract note in
-; a doc comment nobody's data ever exercised.
+; This entry's own PREVIOUS text argued the divergence could not be fixed
+; without a signature change, because [creed]'s inputs (temporal/observed/
+; date) have no notion of "this Mass is a Requiem". That argument was
+; wrong: {!Colour.Black} is used by exactly TWO celebrations in the whole
+; engine (verified by grepping every [colour Black]/[Colour.Black] site in
+; lib/ and data/) -- [commemoration-of-all-souls] (data/ef/sanctoral.sexp)
+; and Good Friday (temporal_ef.ml, already [false] via the RG 23(b) feria
+; exclusion regardless of colour, and has no Mass at all in the
+; 1955-restored Holy Week) -- so on every day this engine can construct,
+; [colour = Black] if and only if the Mass is a Requiem: a citable
+; implication (RG 117 assigns black to Masses of the dead) on THIS DATA,
+; not a heuristic guess. [Rite_ef.Rubrics_ef.creed] now reads
+; [observed.colour] as RG 476(f)'s own guard, positioned ahead of 475(b)'s
+; rank branch; see rubrics_ef.ml's own header and the guard's own comment
+; for the full account, and test_rubrics_ef.ml's
+; [test_colour_black_population_is_exactly_two] for the test that fails
+; loudly the day a third [Colour.Black] celebration is ever added, so this
+; proxy cannot silently rot.
 ;
-; Domain-wide count: unmeasured (out of this task's scope -- would need a
-; sweep of every I-class Requiem-shaped day 1583-9999, which does not
-; exist as a distinct predicate in this codebase to sweep on). The two
-; days this WOULD also apply to in principle -- 2 November itself in any
-; year it does not fall on a Sunday, and the plain All Souls' Day office
-; whenever it is not transferred -- are outside this fixture's own
-; window's reach for a second occurrence, so only the one real day above
-; is witnessed and adjudicated here.
-((id L1)
- (citation "RG 476(f): \"Non dicitur symbolum ... f) in Missis defunctorum\" -- the Creed is never said at a Mass of the dead; rubrics_ef.ml's own header already documents this Mass-type as unmodelled")
- (verdict lms_ordo)
- (note "colitur's [creed] has no notion of a Requiem Mass in its inputs (temporal/observed/date), so RG 475(b)'s I-class branch fires unopposed on All Souls' Day (Class1, black). A pre-existing, self-documented scope limit (rubrics_ef.ml's own header), witnessed here for the first time on a real date rather than left as an abstract note: 2025-11-03, All Souls' Day transferred from Sunday 2 November. The Ordo (\"No Gl Sequence No Cr Pr of the Dead\") is correct on the rubric.")
- (expected_rows 1))
+; BLAST RADIUS, measured (full 1583-9999 sweep of {!Colitur_kernel.
+; Calendar.year}, grouped by [observed]'s own slug): 8 416 days changed
+; from [true] to [false], every one [commemoration-of-all-souls] (missing
+; from exactly 1 of the 8 417 domain years -- a boundary-year transfer
+; effect at 1583/9999, not a third population member: see the "know what
+; each layer cannot see"-style caveats elsewhere in this project for other
+; instances of the same edge). A further 8 416 [Colour.Black] days are
+; Good Friday ([ef-passiontide-2-friday]), unchanged by this fix (already
+; [false] before it, via the earlier feria exclusion) -- confirmed
+; [creed_true_among_black = 0] after the fix, over the WHOLE domain, not
+; merely the fixture's own 2024-2025 window.
+;
+; `colitur day`/`colitur readings` are unaffected (neither prints [creed]);
+; `colitur rubrics` output changes on exactly the 8 416 days above.
+
+; L2 -- OPEN, colitur's own gap, found while building L1's fix (whole-
+; branch review finding 2, 2026-08-22). NOT an active entry in this
+; file's own machine-checked sense (no code loads or cross-checks an
+; "L2" id anywhere -- this file's own [allow_entry] machinery is scoped
+; to the CREED axis, {!test_lms_ordo.test_creed_matches_or_is_explained}
+; alone; a second, unrelated axis reusing the same id-lookup mechanism
+; would let one test's "declared but never fired" check spuriously trip
+; on the other axis's own entry). Recorded here as prose only, the same
+; discipline this file's own closed L1 note above and
+; expected-divergences.sexp's own closed C-entries already use for a
+; citation worth preserving without a live parser hook -- the REAL
+; enforcement is test_lms_ordo.ml's own [test_formulary_override_matches],
+; which asserts this population is exactly 3 days directly.
+;
+; Strengthening test_lms_ordo.ml's formulary-override check from PRESENCE
+; to the actual STRING (this same finding 2) surfaced a real, separate
+; defect in step 3 of the reading chain ({!Rite_ef.Lectionary_ef.
+; readings}), independent of the Creed axis above. Three real days in the
+; fixture window -- 2025-05-30 (Friday), 2025-06-02 (Monday), 2025-06-03
+; (Tuesday), all between Ascension Thursday and the following Sunday --
+; have no proper of their own and fall to {!Colitur_kernel.
+; Mass_formulary.Preceding_sunday}. colitur resumes the nearest CALENDAR
+; Sunday (the "5th"/"6th Sunday after Easter", Paschaltide's own week
+; 6/7), but the Missal, and the Ordo, keep saying ASCENSION'S OWN Mass
+; through this stretch instead: "Mass of the Ascension" on all three
+; days, never a numbered-Sunday title. Step 3 has no notion of "inside
+; Ascension's own after-feast period" -- it always resumes the nearest
+; preceding Sunday, numbered or not.
+;
+; STRUCTURAL, not a fixture artefact: the Ascension is a fixed Easter+39
+; (a Thursday inside Paschaltide week 6), so every feria whose own
+; Preceding_sunday fallback would otherwise resolve to week 6 or 7 falls
+; in this same stretch, every year -- this is not a coincidence of
+; 2024-2025 alone, though only this fixture's own window has been checked
+; against a real Ordo.
+;
+; NOT FIXED in this task: fixing it means teaching step 3, or the
+; lectionary data, about Ascension's own after-feast period -- the same
+; shape as the Christmastide "infra octavam Nativitatis" entries
+; data/ef/lectionary.sexp already carries directly rather than resolving
+; through this fallback -- a lectionary-data change out of a fix round's
+; own scope. test/test_lms_ordo.ml's own
+; [expected_preceding_sunday_override] excludes exactly these two
+; (season, week) pairs from its general mapping, and
+; [test_formulary_override_matches] asserts the excluded population is
+; exactly 3 real days, all showing "Mass of the Ascension", so a change
+; to this population (a fourth day, a different Ordo string) would be
+; caught rather than silently absorbed.
+;
+; Domain-wide count: unmeasured -- "which feria is inside Ascension's own
+; after-feast period" is not a distinct predicate anywhere in this
+; codebase to sweep on. Structurally recurs every year, so on the order
+; of 3 days/year x domain years, not merely 3 total.