fix(lang): stop internal book ids reaching output, and finish the docs

Three defects, all found by running the documented examples rather than
by reading code.

All seven modern-numbering targets carried their own internal id as
their Latin name (kings_1.full = kings_1), so --sigla-tradition modern
printed "kings_1 19:3-8" -- leaking a key book.mli states is never shown
to a reader. They now carry Vulgate-derived Latin names marked
CONSTRUCTED: the 1962 Missal uses Vulgate numbering throughout, so it
can contain no incipit for a book that exists only under a later
convention. In Latin the modern tradition therefore only moves Kings and
Esdras; Osee, Ionas, Ecclesiasticus and the Apocalypse keep their names
either way, since modern numbering is a vernacular convention.

The coverage test could not catch this. It compares a name against its
KEY ("kings_1.full"), and "kings_1" differs from that, so the row
passed. A second check now asserts no name IS its own id. The two are
complementary: one catches a missing row, the other a row present but
filled with the wrong thing. Mutation-proved.

--help still claimed sigla settings were resolved "but not yet
rendered", stale since the render wiring landed. colitur-config.5
documented none of the three settings even though config --show prints
them.

4 files changed, 132 insertions, 17 deletions

diff --git a/bin/main.ml b/bin/main.ml
index 5ad3132..c76a082 100644
--- a/bin/main.ml
+++ b/bin/main.ml
@@ -1267,9 +1267,10 @@ naming:
             `abbr`), the same discipline as an unknown `--lang`; an
             unrecognised `--sigla-tradition` is not -- it degrades to the
             Vulgate with a stderr warning, because asking for a renumbering
-            is optional the way asking for a language is not. Settings
-            only, for now: this build resolves and reports them but does
-            not yet render a citation through them.
+            is optional the way asking for a language is not. `--raw`
+            bypasses all of it and emits each citation exactly as stored,
+            byte for byte, so it stays usable for diffing and does not
+            depend on the citation parser being correct.
 
   colitur lang --list         which language files this build can find, and
                                each one's own declared fallback, if any.
diff --git a/lang/la.ini b/lang/la.ini
index 369d6b3..3783482 100644
--- a/lang/la.ini
+++ b/lang/la.ini
@@ -2157,17 +2157,40 @@ apocalypse.abbr        = Apoc
 ; render one of these bare ids straight into a citation.
 ; UNSOURCED (all seven rows below -- see this section's own header
 ; immediately above for why: a category mismatch, not a search gap).
-kings_1.full           = kings_1
-kings_1.abbr           = kings_1
-kings_2.full           = kings_2
-kings_2.abbr           = kings_2
-nehemiah.full          = nehemiah
-nehemiah.abbr          = nehemiah
-sirach.full            = sirach
-sirach.abbr            = sirach
-hosea.full             = hosea
-hosea.abbr             = hosea
-jonah.full             = jonah
-jonah.abbr             = jonah
-revelation.full        = revelation
-revelation.abbr        = revelation
+; The seven MODERN-NUMBERING targets. All CONSTRUCTED, and necessarily so:
+; the 1962 Missal uses Vulgate numbering throughout, so it can contain no
+; incipit for a book that exists only under a later convention. Latin forms
+; follow the Vulgate's own naming for the same physical book.
+;
+; They were previously set to the internal id ("kings_1.full = kings_1") by
+; applying the UNSOURCED rule mechanically -- but that rule says to fall back
+; to the spelling THE DATA USES, and these books are never cited by the data
+; at all, so the fallback yielded the id. `--sigla-tradition modern` then
+; printed "kings_1 19:3-8", leaking an internal key that book.mli states is
+; never shown to a reader.
+;
+; Note that in LATIN the modern tradition only really moves Kings and Esdras:
+; Osee, Ionas, Ecclesiasticus and the Apocalypse keep their Vulgate names, so
+; those four map to an identical rendering. That is correct, not a no-op bug
+; -- modern numbering is a vernacular convention, and these files show it.
+kings_1.full           = Liber Regum I
+; CONSTRUCTED -- same reasoning as kings_3.full above, one volume lower.
+kings_1.abbr           = 1 Reg
+kings_2.full           = Liber Regum II
+; CONSTRUCTED.
+kings_2.abbr           = 2 Reg
+nehemiah.full          = Liber Nehemiae
+; CONSTRUCTED. The Vulgate calls this book 2 Esdras; see esdras_2 above.
+nehemiah.abbr          = Neh
+sirach.full            = Liber Ecclesiastici
+; CONSTRUCTED. Same book as ecclesiasticus above, under its modern name.
+sirach.abbr            = Sir
+hosea.full             = Osee Propheta
+; CONSTRUCTED. Identical to osee above -- see the note at the top of this block.
+hosea.abbr             = Os
+jonah.full             = Ionas Propheta
+; CONSTRUCTED. Identical to jonas above.
+jonah.abbr             = Ion
+revelation.full        = Apocalypsis
+; CONSTRUCTED. Identical to apocalypse above.
+revelation.abbr        = Apoc
diff --git a/man/colitur-config.5 b/man/colitur-config.5
index 8f21028..6b3a574 100644
--- a/man/colitur-config.5
+++ b/man/colitur-config.5
@@ -181,6 +181,69 @@ accepts it. Relevant only to
 which requires a format either from this setting or from
 .BR \-\-format ;
 no default.
+.TP
+.B sigla_style
+Which citation STYLE to render book names and punctuation with: a language
+code, or a path to a file carrying a
+.B [sigla]
+section. Defaults to the resolved
+.IR lang ,
+so a booklet's citations follow its own language unless told otherwise.
+See
+.BR colitur (1)'s
+.B SIGLA
+section for the five style settings.
+.TP
+.B sigla_book
+.BR full " or " abbr .
+Overrides the selected style's own
+.B book
+setting. Default
+.BR abbr ,
+giving
+.B "Luc. 5:12\-14"
+rather than
+.BR "Evangelium secundum Lucam 5:12\-14" .
+An unrecognised value is a hard error, the same discipline as an unknown
+.BR \-\-lang .
+.TP
+.B sigla_tradition
+A section name in
+.IR lang/traditions.ini ,
+deciding which book a reference DENOTES rather than what it is called.
+Default
+.BR vulgate ,
+the identity \(em colitur never renumbers unless asked. With
+.B modern
+the Vulgate's own
+.B "3 Kings 19:3\-8"
+renders as
+.B "1 Reg 19:3\-8"
+in Latin, or
+.B "1 Kgs 19:3\-8"
+under
+.BR "\-\-lang en" .
+In Latin the modern tradition only really moves Kings and Esdras: Osee,
+Ionas, Ecclesiasticus and the Apocalypse keep their Vulgate names either
+way, because modern numbering is a vernacular convention.
+Unlike
+.IR sigla_book ,
+an unrecognised value is NOT fatal: it warns on stderr and falls back to
+.BR vulgate ,
+because asking for a renumbering is optional the way asking for a language
+is not.
+.PP
+Naming and renumbering are deliberately separate mechanisms.
+.I sigla_style
+selects what a book is CALLED, which varies by language;
+.I sigla_tradition
+selects which book a reference DENOTES, which does not \(em "modern
+numbering" is the same decision in Latin, Polish and English. Conflating
+them is how a citation ends up naming the wrong book.
+.PP
+None of these affect
+.BR \-\-raw ,
+which emits every citation exactly as stored, byte for byte.
 .SH UNKNOWN KEYS AND SECTIONS
 A key inside
 .B [defaults]
diff --git a/test/test_lang_coverage.ml b/test/test_lang_coverage.ml
index b2d32a4..0271a59 100644
--- a/test/test_lang_coverage.ml
+++ b/test/test_lang_coverage.ml
@@ -119,6 +119,32 @@ let test_every_book_named () =
   in
   Alcotest.(check (list string)) "every book named, both forms" [] missing
 
+(* A name must not BE the internal id. The check above compares the value
+   against the KEY ("kings_1.full"), so a row reading `kings_1.full = kings_1`
+   passes it -- the two strings differ. That is exactly what shipped: all seven
+   tradition targets carried their own id as their name, and
+   `--sigla-tradition modern` printed "kings_1 19:3-8", leaking a key that
+   book.mli states is never shown to a reader.
+
+   The two checks are complementary and neither subsumes the other: that one
+   catches a MISSING row, this one catches a row present but filled with the
+   wrong thing. *)
+let test_no_book_name_is_an_internal_id () =
+  let la = la () in
+  let leaked =
+    List.concat_map
+      (fun id ->
+        let n = Colitur_citation.Book.to_string id in
+        List.filter_map
+          (fun form ->
+            let v = L.bible la (n ^ "." ^ form) in
+            if v = n then Some (n ^ "." ^ form) else None)
+          [ "full"; "abbr" ])
+      Colitur_citation.Book.all
+  in
+  Alcotest.(check (list string)) "no book name is its own internal id" []
+    leaked
+
 (* lang/en.ini is DELIBERATELY partial (see its own header note): it declares
    [meta] fallback = la, so a slug it does not carry itself should still
    resolve through the chain to la.ini's name rather than degrade to the bare
@@ -161,4 +187,6 @@ let suite =
       Alcotest.test_case "vocabularies complete" `Quick test_vocabularies_are_complete;
       Alcotest.test_case "every rite has a Latin name" `Quick test_every_rite_has_a_latin_name;
       Alcotest.test_case "every book named, both forms" `Quick test_every_book_named;
+      Alcotest.test_case "no book name is an internal id" `Quick
+        test_no_book_name_is_an_internal_id;
       Alcotest.test_case "en.ini falls back to Latin" `Quick test_en_falls_back_to_latin ] )