Record why H6 reads no title by shape before a family comma

[derek73]

Summary

H6 (a title before a family comma) decided in #701 that only the title vocabulary claims a word before a family comma, never H2's period shape, so Xyz. Smith, John keeps family Xyz. Smith. The decisions entry gave two reasons, the doctrine (the slot is a surname's slot, as at H5) and the absence of need, and named no surname the shape would break. Closing #710 raised the question again ("surnames don't end in periods"). It turns out there is a cost:

• A surname can open with an abbreviated word. Filipino records write Santa and Santo as Sta. and Sto., and H2's shape (two or more letters, one trailing period) claims both.
• Sta. Maria, Juan reads family Sta. Maria today. With the shape rule it would read title Sta., family Maria.
• US Census 2010 surname bearers (tools/name_usage.py, 2026-10-18): Sta. Ana 408, Sta. Maria 376, Sto. Domingo 163, Sta. Cruz 110.
• ste is a particle and no title, so Ste. Marie, Jean never enters H6's run. sta and sto are in no vocabulary set.
• The Census strips punctuation, but Santa Maria is its own entry (12,375 bearers), so these counts are the people recorded with the abbreviation.

H6 stays as decided. This PR records the reason so it isn't asked a third time:

• decisions/H6: a 2026-10-18 entry with the counts and the recompute command. It also records a gap this rule doesn't close: the comma-less Juan Sta. Maria already splits the surname (middle 'Sta.', family 'Maria'). Listing sta/sto as particles fixes that (measured). As never-given particles, they'd make P1's fold take a bare leading Sta (Sta Juan → family 'Sta Juan'). Listed as ambiguous, they keep given 'Sta' and add a report. Neither listing is decided here; that question is Should Juan Sta. Maria read family Sta. Maria? Today it reads middle Sta., family Maria #725.
• rules/H6: the rationale names the abbreviated-surname class, and "Sta. Maria, Juan" → family="Sta. Maria" joins the boundary examples, so the reading is now an executable pin.
• corpus_rules.jsonl and snapshot.jsonl each gain the one new example name. No other reading moves.

No parser code changes.

Review

The design-docs review found four inaccuracies in the first draft of the decisions entry, all fixed in 7d8be67: the Juan Sta. Maria example, the Ste. reason, how to read the Census counts, and the unmeasured Mt./Ft. counts. Its review of that fix commit came back clean, and ba7a3ba states the one condition it flagged.

Test plan

• uv run pytest: 21,314 passed
• The snapshot diff is the one new name (Sta. Maria, Juan)

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.04%. Comparing base (16ee285) to head (ba7a3ba).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #724   +/-   ##
=======================================
  Coverage   99.04%   99.04%           
=======================================
  Files          46       46           
  Lines        4585     4585           
=======================================
  Hits         4541     4541           
  Misses         44       44           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.