fix(ef): M6 was convicted on the transcription's silence, not a scan

data/ef/expected-divergences-missalemeum.sexp's M6 entry and its matching
comment in test/test_oracle.ml both cited 1962-06-23,_SS_Ioannes_XXIII,
_Missale_Romanum,_LT.pdf's 14 August row ("Vigilia, II classis." --
nothing else) as if it were the calendarium itself. It is an electronic
transcription that silently drops commemoration lines the photographic
scans carry. missale-romanum-1962.pdf's 14 August row reads "Vigilia, II
classis, Commemoratio S. Eusebii Conf.", with its own proper following --
independently confirmed in the second scan, Missale Romanum 1962_text.pdf.

M6's verdict flips colitur -> missalemeum: this is a genuine colitur data
gap, the same shape as M4/M7/M9/M14 (absent from data/ef/sanctoral.sexp and
from lectio's own tridentine-calendar.ini), not a spurious oracle entry.
Not fixed here for the same reason those four are not: it requires
touching lectio's own bootstrap source, out of reach for this task.

expected_rows is unchanged (2) -- the diff shape the allow-list predicate
matches does not depend on which side is judged correct, confirmed by the
suite passing unchanged.

A companion register ruling that convicted romanus (9 August) on the
identical reasoning -- and the register's own sanctoral-audit bucket
counts that followed from it -- are corrected in docs/research/
rules-register.md (gitignored, not part of this commit). Full audit
method, the complete list of dropped commemorations found across the
whole calendarium, and every other ruling checked for the same defect
are in .superpowers/sdd/2026-08-12-colitur-rg16a/transcription-audit-
report.md (also gitignored).

2 files changed, 28 insertions, 16 deletions

diff --git a/data/ef/expected-divergences-missalemeum.sexp b/data/ef/expected-divergences-missalemeum.sexp
index f50898b..b5a9508 100644
--- a/data/ef/expected-divergences-missalemeum.sexp
+++ b/data/ef/expected-divergences-missalemeum.sexp
@@ -59,9 +59,9 @@
  (note "Independently confirmed by the oracle, not a new finding: missalemeum's \"Pro rogationibus\" commemoration on St Mark's day (25 April, the Major Litanies' fixed date) is exactly the missing office register §6 already tracks. Only 2026 shows as a comparator MISMATCH: 25 April 2026 is a Saturday with Mark winning outright, so colitur's commemoration list is empty where the oracle's carries \"Pro rogationibus\" -- a real presence gap. 25 April 2027 is a Sunday: the Sunday wins on BOTH sides, and each side's own single commemoration slot goes to a DIFFERENT candidate (colitur: St Mark himself, an ordinary II-class sanctoral loser admitted under RG111(b)'s \"de festo II classis\" slot; missalemeum: the Major Litanies) -- both non-empty, same count, so this comparator's count-only axis (deliberately not slug-identity, see test_oracle.ml's own header) cannot see that mismatch; the underlying gap is the same, just invisible to this year's row.")
  (expected_rows 1))
  ((id M6)
- (citation "1962 calendarium, 14 August row: \"Vigilia, II classis.\" -- nothing else listed -- primary-source-verified 2026-08-12")
- (verdict colitur)
- (note "The reverse of M4/M7/M9: no commemoration is listed for 14 August anywhere in the primary source this project treats as authoritative. data/ef/sanctoral.sexp agrees (a single entry, vigil-of-the-assumption). missalemeum's \"St. Eusebius\" for this date has no support found here -- a DIFFERENT St Eusebius (bishop and martyr) is genuinely commemorated 16 December, matching lectio's own [eusebius] entry at THAT date, not this one.")
+ (citation "CORRECTED 2026-08-12, transcription-audit task (branch ef-rg16a): the previous citation below read the ELECTRONIC TRANSCRIPTION (1962-06-23,_SS_Ioannes_XXIII,_Missale_Romanum,_LT.pdf), which silently drops calendarium commemorations, and mistook its silence for the calendarium's own. A photographic scan (missale-romanum-1962.pdf), 14 August row: \"Vigilia, II classis, Commemoratio S. Eusebii Conf.\", with its own proper immediately following (\"Eodem die ... Et fit commemoratio S. Eusebii Conf.\") -- independently confirmed in the second scan, Missale Romanum 1962_text.pdf. Primary-source-verified 2026-08-12.")
+ (verdict missalemeum)
+ (note "The same shape as M4/M7/M9/M14, not the reverse of them: a genuine commemoration -- S. Eusebius, Confessor, 14 August -- is present in the 1962 calendarium and in missalemeum's own data, but absent from data/ef/sanctoral.sexp and from lectio's own tridentine-calendar.ini (no 08-14 entry beyond vigil-of-the-assumption). This Eusebius Confessor (14 Aug) is a genuinely DIFFERENT saint from Eusebius Bishop and Martyr (16 Dec, already correctly in data/ef/sanctoral.sexp, matching lectio's own [eusebius] entry at THAT date) -- both exist, at their own separate dates; the calendarium carries both. Not fixed here: single-source data, the same bootstrap-reach constraint as M4/M7/M9/M14 (would require touching ~/git/projects/lectio). Register docs/research/rules-register.md, Decisions item 4 and the transcription-audit record in section 6, has the full account and the wider calendarium sweep this correction came from.")
  (expected_rows 2))
  ((id M7)
  (citation "1962 calendarium, 26 October row: \"Com. S. Evaristi Papæ et Mart.\" -- primary-source-verified 2026-08-12")
diff --git a/test/test_oracle.ml b/test/test_oracle.ml
index ac8d95d..0aed1b9 100644
--- a/test/test_oracle.ml
+++ b/test/test_oracle.ml
@@ -345,19 +345,31 @@ let m4_dates = [ "2026-01-28"; "2027-01-28" ]
    report). *)
 let m5_dates = [ "2026-04-25" ]
 
-(* M6 -- the reverse of M4/M7/M9: the primary calendarium's own row for 14
-   August reads, in full, "Vigilia, II classis." -- nothing else. No
-   commemoration is listed for that date anywhere in the source this
-   project treats as authoritative (CLAUDE.md binding decision 1). data/ef/
-   sanctoral.sexp agrees (a single entry, [vigil-of-the-assumption], 14
-   Aug). missalemeum's own "St. Eusebius" commemoration for that date has
-   no support in the 1962 universal calendar found here (a DIFFERENT St
-   Eusebius, bishop and martyr, is genuinely commemorated 16 December,
-   matching lectio's own [eusebius] entry at that date -- not this one).
-   Verdict colitur: this is the one direction among M4/M6/M7/M9 where the
-   oracle itself appears to be carrying data outside the 1962 universal
-   calendar (a pre-1955 survival or a different edition), not colitur
-   missing something real. *)
+(* M6 -- CORRECTED 2026-08-12, transcription-audit task (branch ef-rg16a):
+   the previous version of this comment read the ELECTRONIC TRANSCRIPTION
+   (1962-06-23,_SS_Ioannes_XXIII,_Missale_Romanum,_LT.pdf), which silently
+   drops calendarium commemorations, and mistook its silence for the
+   calendarium's own -- "reads, in full, 'Vigilia, II classis.' -- nothing
+   else" was true of the TRANSCRIPTION only. A photographic scan
+   (missale-romanum-1962.pdf), 14 August row: "Vigilia, II classis,
+   Commemoratio S. Eusebii Conf.", with its own proper immediately
+   following ("Eodem die ... Et fit commemoratio S. Eusebii Conf.") --
+   independently confirmed in the second scan, Missale Romanum 1962_text
+   .pdf. The same shape as M4/M7/M9/M14, NOT their reverse: a genuine
+   commemoration is present in the calendarium and in missalemeum's own
+   data, but absent from data/ef/sanctoral.sexp and lectio's own
+   tridentine-calendar.ini (no 08-14 entry beyond [vigil-of-the-assumption]).
+   This Eusebius Confessor (14 Aug) is a genuinely DIFFERENT saint from
+   Eusebius Bishop and Martyr (16 Dec, already correctly in data/ef/
+   sanctoral.sexp, matching lectio's own [eusebius] entry at THAT date) --
+   both exist, at their own separate dates. Verdict missalemeum (flipped
+   from colitur): colitur is missing this commemoration, a genuine data gap
+   not fixable without touching ~/git/projects/lectio -- see
+   docs/research/rules-register.md, Decisions item 4 and the
+   transcription-audit record in section 6, for the full account and the
+   wider calendarium sweep this correction came from. The diff SHAPE this
+   predicate matches (Comm_presence, both dates) is unchanged by the
+   verdict flip -- only which side is right, and why, changed. *)
 let m6_dates = [ "2026-08-14"; "2027-08-14" ]
 
 (* M7 -- a genuine DATA GAP, primary-source-confirmed, the same shape as