docs: the project-ordinal headlines leave the coverage record too (#737)

35 sites — 19 in docs/language-maturity.md, 16 in catalogue.tsv, which is
hand-maintained rather than generated (gen-seed.py carries none of them).

Same rule as the code sweep: the typology survives, the ordinal goes. "the
fleet's FIRST Bodish/Tibetic branch" → "Bodish/Tibetic (Sino-Tibetan)", "the
fleet's 4th creole (after ht/kea/pcm)" → "a creole".

KEPT, because they are comparisons rather than headlines — each names its
comparand or gives the number:
  - "one of the fleet's thinnest anchors (7 words vs Luo 17 / Madurese 35)"
  - "one of the fleet's LARGEST referees (63,024 human headwords)"
  - "like the fleet's other ejective languages Quechua/Georgian"
  - "the fleet's Iranian modules (fa/ps/ckb) don't cover"
A reader can check every one of those; "the fleet's FIRST X" is unfalsifiable
without the commit history.

NOT TOUCHED: docs/investigations/*.md, which carry ~18 more. Those are the
defence thesis — a bring-up log is exactly where "this was the first Mayan
language" belongs, and the export deletes the directory anyway.

Verified the doc→tool contract survives: masking-coefficient.ts still parses
the verdict column out of language-maturity.md, so the table structure is
intact.

227 files / 3129 tests unchanged.


Claude-Session: https://claude.ai/code/session_012Lc3WnUgogC7okV7n53vjr

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: christopherthompson81

README.md

@@ -20,19 +20,27 @@ import { phonemizeAsync } from "vernacula-phonemizer";
 // Worst input, best output. A language written in two scripts yields ONE canonical IPA either way (Tashelhit in
 // Latin/Tifinagh, Fula's Adlam, Bambara's N'Ko, Sundanese's Aksara Sunda, Zhuang's Sawndip).
 await phonemizeAsync("I read a book", "en"); // Latin → aᶦ ɹˈɛd ə bˈʊk
-await phonemizeAsync("Taclḥit", "shi"); // Berber Latin → taʃlħit
+await phonemizeAsync("世界", "cmn"); // Han → ʂʐ̩˥˩ t͡ɕiɛ˥˩
+await phonemizeAsync("भारत", "hi"); // Devanagari → bʱˈaːɾət̪
+await phonemizeAsync("العربية", "ar"); // Arabic → alʕarabˈijːa
+await phonemizeAsync("Україна", "uk"); // Cyrillic → ukrajina
+await phonemizeAsync("বাংলাদেশ", "bn"); // Bengali → baŋlad̪eʃ
+await phonemizeAsync("ひらがな", "ja"); // Hiragana → çiɾäɡäꜜnä
+await phonemizeAsync("カタカナ", "ja"); // Katakana → kätäkänä
+await phonemizeAsync("日本語", "ja"); // Kanji → niho̞ŋɡo̞
+
 await phonemizeAsync("ⵜⴰⵛⵍⵃⵉⵜ", "shi"); // Tifinagh → taʃlħit
 await phonemizeAsync("𞤆𞤵𞤤𞤢𞥄𞤪", "ff"); // Adlam → pˈulaːɾ
 await phonemizeAsync("ߓߊߡߊߣߊ߲", "bm"); // N'Ko → bamanã
 await phonemizeAsync("Ελληνικά", "el"); // Greek → elinika
-await phonemizeAsync("Україна", "uk"); // Cyrillic → ukrajina
+
 await phonemizeAsync("Հայերեն", "hy"); // Armenian → hɑjeɾen
 await phonemizeAsync("ქართული", "ka"); // Georgian → kʰaɾtʰuli
 await phonemizeAsync("עברית", "he"); // Hebrew → ʔivʁit
-await phonemizeAsync("भारत", "hi"); // Devanagari → bʱˈaːɾət̪
+
 await phonemizeAsync("ਪੰਜਾਬੀ", "pa"); // Gurmukhi → pˈə̃ɲd͡ʒaːbiː
 await phonemizeAsync("ગુજરાતી", "gu"); // Gujarati → ɡˈud͡ʒɾat̪i
-await phonemizeAsync("বাংলাদেশ", "bn"); // Bengali → baŋlad̪eʃ
+
 await phonemizeAsync("ꠍꠤꠟꠐꠤ", "syl"); // Syloti Nagri → silʈi
 await phonemizeAsync("ଓଡ଼ିଆ", "or"); // Odia → ˈoɽia
 await phonemizeAsync("தமிழ்", "ta"); // Tamil → t̪ˈɐmɪɻ
@@ -41,17 +49,15 @@ await phonemizeAsync("ಕನ್ನಡ", "kn"); // Kannada → kˈanːaɖa
 await phonemizeAsync("മലയാളം", "ml"); // Malayalam → mˈalajaːɭam
 await phonemizeAsync("සිංහල", "si"); // Sinhala → sˈiŋhələ
 await phonemizeAsync("ᱥᱟᱱᱛᱟᱲᱤ", "sat"); // Ol Chiki → santaɽi
-await phonemizeAsync("العربية", "ar"); // Arabic → alʕarabˈijːa
+
 await phonemizeAsync("فارسی", "fa"); // Perso-Arabic → faːɾsˈiː
 await phonemizeAsync("አማርኛ", "am"); // Geʽez → amaɾɲa
 await phonemizeAsync("မြန်မာ", "my"); // Myanmar → mja˨ɴma˨
 await phonemizeAsync("လိၵ်ႈတႆး", "shn"); // Shan → lik̚˧˧˨taj˥
 await phonemizeAsync("ខ្មែរ", "km"); // Khmer → kʰmae
 await phonemizeAsync("ภาษาไทย", "th"); // Thai → pʰˈaː˧saː˩˩˦tʰˌa˧j
-await phonemizeAsync("世界", "cmn"); // Han → ʂʐ̩˥˩ t͡ɕiɛ˥˩
-await phonemizeAsync("ひらがな", "ja"); // Hiragana → çiɾäɡäꜜnä
-await phonemizeAsync("カタカナ", "ja"); // Katakana → kätäkänä
-await phonemizeAsync("日本語", "ja"); // Kanji → niho̞ŋɡo̞
+
+
 await phonemizeAsync("한국어", "ko"); // Hangul → hˈɐnɡuɡɘ
 await phonemizeAsync("ꦗꦮ", "jv"); // Javanese → d͡ʒˈɔwɔ
 await phonemizeAsync("ᮞᮥᮔ᮪ᮓ", "su"); // Aksara Sunda → sˈunda