rite(ef): fix false justification in impedes_vigil's comment

Review finding: the comment claimed every RG 91 entry that can outrank a
II-class vigil (entry 21) without being a Sunday is I class 'by the table's
own structure (entries 1-13)'. False -- entries 14 and 16-20 (Feasts of the
Lord II class, universal/proper/indult II-class feasts, days within the
Nativity octave) are all Class2, all outrank entry 21, and none is a Sunday.

The code was always correct: impedes_vigil implements RG 33's own two named
conditions (any Sunday, or a I-class feast) directly, and does not depend on
the band table's numeric ordering at all. Reworded to say so, citing the
counter-example entries the review named instead of appealing to a table
structure that does not guarantee what the old comment claimed.

Comment-only change; no logic, signature, or test changes.

1 files changed, 11 insertions, 7 deletions

diff --git a/lib/rites/rite_ef/precedence_ef.ml b/lib/rites/rite_ef/precedence_ef.ml
index d8d64a8..02181ed 100644
--- a/lib/rites/rite_ef/precedence_ef.ml
+++ b/lib/rites/rite_ef/precedence_ef.ml
@@ -243,13 +243,17 @@ let contains_substring s ~needle =
 
 let is_sunday_slug slug = contains_substring slug ~needle:sunday_marker
 
-(* RG 33's "any Sunday or a I-class feast": every RG 91 entry that can ever
-   outrank a II-class vigil (entry 21) without itself being a Sunday is I
-   class by the table's own structure (entries 1-13), so [rank = Class1]
-   alone already covers every way a I-class vigil (entries 5, 9 -- Nativity,
-   Pentecost) can be impeded at all; the slug check is what a II-class vigil
-   impeded by an ordinary Sunday (entry 15, rank Class2) needs, since that
-   winner's own rank is not Class1. *)
+(* RG 33's own two conditions, taken directly from its text ("any Sunday or
+   a I-class feast") -- NOT derived from anything about which RG 91 entries
+   can numerically outrank a vigil. [rank = Class1] is the "I-class feast"
+   half. [is_sunday_slug] is the "any Sunday" half, and it is not redundant
+   with the rank check: RG 91 entries 14 and 16-20 (Feasts of the Lord II
+   class, universal/proper/indult II-class feasts, days within the Nativity
+   octave) are all [Class2], all outrank a II-class vigil (entry 21), and
+   none of them is a Sunday -- a winner of that shape satisfies neither
+   condition here, so [impedes_vigil] correctly returns [false] and such a
+   vigil falls through to RG 95's ordinary commemorate-or-omit branch
+   instead of RG 33's omission, exactly as the rubric requires. *)
 let impedes_vigil (winner : Vocab_ef.rank Precedence.candidate) =
   let cel = winner.Precedence.cel in
   cel.Celebration.rank = Vocab_ef.Class1