Merge branch 'ef-rebootstrap': re-bootstrap the sanctoral, refresh the differential

Re-bootstraps data/ef/sanctoral.sexp from lectio's corrected generator
(322 -> 327 entries), propagating fixes colitur had inherited wholesale:
15 entries wrongly marked commemoration-only that are really III-class
feasts, four entries missing entirely, six commemorations an electronic
transcription had hidden, and two class corrections.

38639 days change across 1583-9999, in three shapes with no remainder:
2245 where a III-class feast now correctly beats a class-4 feria, 35174
where one of five restored commemorations appears, and 1220 on 13
January. All five restored entries were verified against a photographic
scan -- real, correctly ranked, and confined to their own dates.

13 January is NOT fixed by this. It moves from one wrong answer to a
different one, because RG 112 and a Holy Family office are both unbuilt.
Recorded as its own allow-list entry with a citation and a count, rather
than left inside a season-boundary entry's cardinality pin, and
deliberately not pinned by a golden test.

Then refreshes the lectio differential fixture, which predated the seven
EF defects fixed in that project. Nine allow-list entries close -- each
was a real divergence adjudicated in colitur's favour, and each closes
because lectio was fixed, with its RG citation preserved. The list goes
from twelve entries to four; the remaining 540 differences are fully
attributed with none unexplained.

1 files changed, 72 insertions, 1 deletions

diff --git a/data/ef/adjustments.sexp b/data/ef/adjustments.sexp
index 05ee401..e33b4eb 100644
--- a/data/ef/adjustments.sexp
+++ b/data/ef/adjustments.sexp
@@ -50,7 +50,78 @@
 ; the user's ruling resolves that disagreement in the oracle's favour for
 ; this codebase's own purposes, not by declaring the calendarium argument
 ; wrong. See register §6.0 for the full account of both sides.
+;
+; ef-rebootstrap (2026-08-12): both directives re-examined against the
+; regenerated source (tridentine-calendar.ini, SHA-256 1b303ef2...), which
+; independently fixed several of the same defects this file was written to
+; patch. Neither directive was removed -- reasoning below, per the task's
+; own instruction to decide deliberately rather than delete on sight.
+;
+; `Edit most-holy-name-of-mary` -- the base bootstrap no longer needs this
+; correction: the regenerated INI's own `[most-holy-name-of-mary]` section
+; has DROPPED its `class = lord` field entirely (confirmed by direct grep
+; against the source), so `tools/bootstrap_sanctoral.ml`'s `parse_subject`
+; now falls through to its own default, `Subject.Saint` -- not `Lord` --
+; before this overlay ever runs. Checked what {!Overlay.apply}'s [Edit]
+; actually does with a directive whose target slug is present but whose
+; field is no longer wrong, rather than assuming: [Edit] has no notion of
+; "already correct" -- it looks up the slug, and if found (as this one
+; still is) unconditionally folds every field_edit over the current value,
+; firing NO diagnostic either way (overlay.ml's [apply_directive], the
+; [Edit] branch: the only diagnostic is "slug not present", never "already
+; matches"). So this is not a no-op: it still forces `Saint -> Bvm` on
+; every run, identically to before, just starting from a different base
+; value than it used to. Decision: KEEP, for two independent reasons, not
+; one -- (1) DATA PRECISION, unaffected by the base bootstrap's own fix:
+; the calendarium's "Sanctissimi Nominis Mariae" title names Mary
+; specifically, so [Bvm] remains the more accurate tag than the generic
+; [Saint] default, on the same textual grounds as the original entry
+; above, regardless of what lectio's own `class` field happens to say this
+; week. (2) REGRESSION DEFENCE: because [Set_subject] is unconditional, it
+; also now stands as a guard against `class = lord` ever being
+; RE-introduced for this slug by a future lectio regeneration -- the
+; overlay would still force the result away from [Lord] (the one subject
+; value {!Precedence_ef.band}'s RG 16(a) branch behaviourally reads),
+; rather than silently letting a re-introduced bootstrap defect through.
+; Removing the directive now would trade a currently-harmless redundancy
+; for the loss of that guard.
+;
+; `Suppress vigil-of-christmas` -- re-verified, not assumed: `[vigil-of-
+; christmas]` (24 Dec) is still present, unchanged, in both the
+; regenerated INI and the regenerated `data/ef/sanctoral.sexp` (still the
+; same duplicate of `rite_ef/temporal_ef.ml`'s own `ef-nativity-vigil`
+; this directive was written to remove), so `Layer.mem` still finds it and
+; {!Overlay.apply}'s [Suppress] branch still fires its ordinary
+; slug-present path (no diagnostic) exactly as before. Nothing about this
+; regeneration touched the reason this directive exists; kept unchanged.
+;
+; `Edit eusebius-confessor` -- NEW (ef-rebootstrap fix round 1, F3): this
+; new entry's own bootstrapped colour, `Red`, is a LECTIO DATA DEFECT,
+; found by checking the primary scan directly rather than trusting lectio's
+; ini or missalemeum's own `:r` colour tag, which agree with each other but
+; not with the Missal -- exactly the "two non-primary witnesses agreeing
+; against the primary text" trap this project's own transcription-audit
+; task already named once. `missale-romanum-1962.pdf`, 14 August, the
+; commemoration's own proper: *"Eodem die 14 augusti / S. Eusebii Conf. /
+; Commemoratio / Missa Iustus, ut in festo S. Pauli primi Eremitae, die 15
+; ianuarii..."* -- "Conf." (Confessor, not Martyr), and the borrowed Mass
+; "Iustus" is the SAME Mass 15 January's own St Paul the First Hermit uses
+; -- a Confessor, white in both lectio's own data and colitur's
+; (`paul-the-first-hermit`, `data/ef/sanctoral.sexp`) -- for the Common of
+; Confessors. Corroborated negatively: lectio gives WHITE to every other
+; Confessor/Abbot commemoration in its own data (`maur-abbot`, `giles`,
+; `remigius`, `didacus`, `ubaldus`, `hilarion`, `sabbas`, `silvester`,
+; `alexis`) -- Eusebius Confessor is the one exception, with no textual
+; support found for it. No behavioural impact today (`Commemoration_only`
+; entries are never the OBSERVED celebration, so this colour is never
+; printed by the current pipeline -- but `Record`/rendering will read it
+; the moment a template does), fixed for the data's own accuracy via the
+; SAME overlay mechanism as the other two directives, since
+; `data/ef/sanctoral.sexp` is generated and the source is upstream. Logged
+; as a lectio-side generator/data defect in docs/research/rules-register.md
+; for a future generator fix, not merely patched here silently.
 ((id ef-adjustments)
   (directives
     ((Suppress vigil-of-christmas)
-      (Edit most-holy-name-of-mary ((Set_subject Bvm))))))
+      (Edit most-holy-name-of-mary ((Set_subject Bvm)))
+      (Edit eusebius-confessor ((Set_colour White))))))