feat(ef): implement the calendarium's bissextile February shift

The Missale Romanum's calendarium footnote for February (LT.txt:5011-5014,
scan-corroborated scan2.txt:3050-3058): in a leap year the sixth kalends of
March (24 February) is doubled rather than a 29th day appended, so St
Matthias moves 24->25 February and St Gabriel of Our Lady of Sorrows
27->28, with 24 February itself carrying no fixed office that year.
data/ef/expected-divergences-lms.sexp entry L3 tracked this as an open gap.

Implemented as the rubric's own general mechanism (every fixed entry from
24 through 28 February shifts one civil day later in a leap year), not as
"move these two named saints": the two readings are indistinguishable on
shipped data (nothing else is fixed in that window) and diverge only for a
future --overlay entry in the same range, which the mechanism reading
covers correctly and a two-saints special case would not. This project has
already been bitten once (RG 16(a)) by a rule implemented against shipped
data's coincidental shape rather than the rubric itself.

Kernel stays rite-agnostic: Rite.t gains a fixed_key field (Date.t -> (int
* int) option), the same seam easter already establishes, read only by
Layer.on_date's FIXED half -- the movable half (Easter_offset/Nth_weekday)
is untouched. Default is the identity mapping, an optional argument on
on_date, so every existing caller and every rite that supplies nothing is
byte-identical to before this field existed. Date.is_leap is exposed in
date.mli (already existed in date.ml) so the rite reads the kernel's own
single-sourced leap-year definition. The EF implementation lives in
Rite_ef.Temporal_ef.bissextile_fixed_key, fully cited, wired into
Rite_ef.context.

Blast radius, measured over the full 1583-9999 domain (Calendar.year run
twice per civil year, real fixed_key vs identity, every day diffed): all
2,041 leap years in the domain show a visible change for each saint;
6,983 individual liturgical days change total, zero unclassified, across
four shapes (Matthias vacates 24 Feb in 1,803 years / occupies 25 Feb in
1,811; Gabriel vacates 27 Feb in 1,699 / occupies 28 Feb in 1,670).
L3's own prior estimate (1,650 years, observed-outright only) is confirmed
close on the same definition (1,677 measured); the broader observed-or-
commemorated count is higher, not lower, showing the narrow estimate
undercounted visible impact rather than overcounting it.

The lectio differential (2005-2050 fixture) gains a new cited entry, C40
in data/ef/expected-divergences.sexp: lectio implements no such shift, so
every leap year in its window now diverges on the two shifted days (21
rows, not the naive 44 -- the comparator never compares commemorations, so
a side where the shifted saint has zero comparator-visible footprint on
both engines produces no row). The LMS Ordo's 2023-2024 fixture -- the
window L3 was originally found in -- now matches on the Creed comparison;
its formulary-override bucket count is corrected 182->181 (24 February
leaves the counted Proper population, becoming Own_slug-sourced). The
missalemeum oracle fixtures (2026-2027, 2038, 2035) cover no leap year and
are unaffected.

L3 is closed: converted from an active allow-list record to a prose
closure paragraph (L1's own precedent), citation preserved verbatim, fix
and measured blast radius recorded. The id-list assertion narrows to [L4]
alone, and the now-dead "2024-02-24" -> "L3" date mapping is removed.

Six golden values pinned across four new test cases: Matthias in a leap
year (both the vacated and occupied sides) and a common-year control;
Gabriel likewise, deliberately choosing a leap year where he is admitted
only as an ordinary commemoration rather than winning outright, a
different shape from Matthias's.

Mutation-tested: forcing fixed_key back to identity reddens exactly five
test cases -- the differential's Layer C count-pin (C40 drops to 0 actual
vs 21 expected), the LMS Ordo's Creed and formulary-bucket checks, and
both new golden pins -- confirming the tests actually exercise the fix.
Reverted; dune test, the exhaustive sweep, and make check are all green.

3 files changed, 77 insertions, 0 deletions

diff --git a/lib/rites/rite_ef/rite_ef.ml b/lib/rites/rite_ef/rite_ef.ml
index e48ff33..ba1274b 100644
--- a/lib/rites/rite_ef/rite_ef.ml
+++ b/lib/rites/rite_ef/rite_ef.ml
@@ -39,6 +39,10 @@ let context ~lectionary ~commons : (Vocab_ef.season, Vocab_ef.rank) Rite.t =
        Julian-reckoning rite can supply {!Computus.julian_easter}
        instead. Read by [Layer.index] for [Date_spec.Easter_offset]. *)
     easter = Colitur_kernel.Computus.gregorian_easter;
+    (* The calendarium's own bissextile footnote (February, LT.txt:5011-5014):
+       see {!Temporal_ef.bissextile_fixed_key} for the full citation and
+       mechanism. Read by [Layer.on_date]'s fixed half only. *)
+    fixed_key = Temporal_ef.bissextile_fixed_key;
     rules =
       { Precedence.band = Precedence_ef.band;
         disposition = Precedence_ef.disposition;
diff --git a/lib/rites/rite_ef/temporal_ef.ml b/lib/rites/rite_ef/temporal_ef.ml
index c16732b..1c916f4 100644
--- a/lib/rites/rite_ef/temporal_ef.ml
+++ b/lib/rites/rite_ef/temporal_ef.ml
@@ -1158,6 +1158,68 @@ let anchors y =
      | Some d -> [ ("ef-holy-name-sunday", d) ]
      | None -> [ ("ef-holy-name", holy_name_fallback_date y) ])
 
+(* The Missale Romanum's own CALENDARIUM table, February, footnote
+   (docs/research/LT.txt:5011-5014, scan-corroborated at
+   docs/research/scan2.txt:3050-3058, where OCR mangles it to
+   "bi&sextili"/"Matthue"/"Cabrielis" -- transcription and scan agree, so
+   this rests on two witnesses of the same document):
+
+     "In anno bissextili mensis februarius est dierum 29, et festum S.
+      Matthiae celebratur die 25 februarii, ac festum S. Gabrielis a
+      Virgine perdolente 28 februarii, et bis dicitur sexto calendas, id
+      est die 24 et die 25; et littera dominicalis, quae assumpta fuit in
+      mense ianuario, mutetur in praecedentem..."
+
+   In a leap year February has 29 days, St Matthias is kept on 25 February
+   (not 24), St Gabriel of Our Lady of Sorrows on 28 February (not 27), and
+   the sixth kalends of March is said TWICE -- "bis dicitur sexto
+   calendas... die 24 et die 25". That last clause names the MECHANISM, not
+   two independent exceptions: the Roman calendar's intercalary day is
+   inserted by DOUBLING the sixth kalends (civil 24 February in a common
+   year), not by appending a 29th day at the month's end. Reckoned by
+   kalends position: 24 Feb = VI Kal. Mart., 25 Feb = V Kal., 26 Feb = IV
+   Kal., 27 Feb = III Kal., 28 Feb = pridie Kal. In a leap year VI Kal.
+   itself falls on TWO consecutive civil days (24th and 25th, "bis"), and
+   every kalends position after it is pushed one civil day later as a
+   result: V Kal. from 25th to 26th, IV Kal. from 26th to 27th, III Kal.
+   (Gabriel) from 27th to 28th, pridie Kal. from 28th to 29th. A feast fixed
+   at VI Kal. itself (Matthias) is kept on the SECOND occurrence, not the
+   first -- 25 February carries it, 24 February carries no fixed office at
+   all that year (confirmed against the LMS Ordo witness,
+   data/ef/expected-divergences-lms.sexp entry L3: 24 February 2024 is
+   titled plainly "EMBER SATURDAY of LENT", no Matthias anywhere near it).
+
+   IMPLEMENTED AS THE GENERAL MECHANISM, not as "shift these two named
+   feasts": only three sanctoral entries exist in 23-29 February on shipped
+   data (peter-damien the 23rd, before the doubled kalends and therefore
+   unaffected; matthias the 24th; gabriel-of-our-lady-of-sorrows the 27th),
+   so "shift the two named feasts" and "shift every fixed entry from 24
+   February" are indistinguishable on universal data -- they produce
+   identical output. They differ only for a locally-supplied entry, e.g. an
+   [--overlay] patronal feast fixed at 26 February: under the mechanism
+   reading it shifts to 27 February in a leap year (IV Kal. -> the civil day
+   IV Kal. falls on that year), exactly as a diocesan calendar compiled
+   against the same kalends convention would expect. The rubric's own text
+   states a MECHANISM ("bis dicitur sexto calendas"), not a closed list of
+   two saints, and this project has already been bitten once by a rule
+   implemented against the shipped data's coincidental shape rather than the
+   rubric itself (RG 16(a), CLAUDE.md's own carried lesson) -- so the general
+   reading is the one implemented here.
+
+   Threaded through {!Rite.t}'s [fixed_key] field, read only by
+   {!Colitur_kernel.Layer.on_date}'s FIXED half: the kernel stays
+   rite-agnostic (a Byzantine or other non-Roman rite supplies nothing and
+   is unaffected), and the MOVABLE half ({!Date_spec.Easter_offset},
+   {!Date_spec.Nth_weekday}) is untouched -- nothing in the footnote
+   concerns Easter-relative dates. *)
+let bissextile_fixed_key (d : Date.t) : (int * int) option =
+  let m = Date.month d and day = Date.day d in
+  if m = 2 && Date.is_leap (Date.year d) then
+    if day = 24 then None (* the FIRST VI Kal.: no fixed entry lands here in a leap year *)
+    else if day >= 25 && day <= 29 then Some (2, day - 1) (* the SECOND VI Kal. onward, one day later *)
+    else Some (m, day)
+  else Some (m, day)
+
 (* Compile-time check that this module satisfies the kernel's rite contract. *)
 module _ : Colitur_kernel.Temporal.RITE = struct
   let id = id
diff --git a/lib/rites/rite_ef/temporal_ef.mli b/lib/rites/rite_ef/temporal_ef.mli
index 5e85111..2bed5b6 100644
--- a/lib/rites/rite_ef/temporal_ef.mli
+++ b/lib/rites/rite_ef/temporal_ef.mli
@@ -51,6 +51,17 @@ val week : Date.t -> int option
 (** The lectionary key for a Sunday, or [None] if [d] is not a Sunday. *)
 val sunday_slug : Date.t -> string option
 
+(** The calendarium's own bissextile (leap-year) footnote, February
+    (docs/research/LT.txt:5011-5014): in a leap year the sixth kalends of
+    March (24 February) is doubled rather than a 29th day appended, so every
+    fixed entry from 24 through 28 February is kept one civil day later than
+    usual (Matthias 24->25, Gabriel of Our Lady of Sorrows 27->28), and 24
+    February itself carries no fixed office that year. Supplied as {!Rite.t}'s
+    [fixed_key] field -- see that field's own doc comment and this function's
+    definition for the full citation and the general-mechanism-vs-two-named-
+    feasts reasoning. *)
+val bissextile_fixed_key : Date.t -> (int * int) option
+
 val id : string
 
 (** Total over 1583..9999: every date yields exactly one temporal identity. *)