test(validation): cover @vocab suppressing the coverage finding

[LukasGold]

Closes the one demand in #120 that had no test behind it, and corrects two comments the context.coverage split left stale.

Changes

• tests/data/oold/broken/vocab_covers_the_remainder.schema.json - new fixture. Byte-identical to missing_context_term.schema.json apart from $id, title, and one added @vocab, so the two are each other's control.
• test_vocab_suppresses_the_coverage_finding - asserts context.coverage stays OK, and that context.predicates reports mapped: 2, which is what proves @vocab mapped orphan rather than the check simply standing down.
• check_registry.py - the comment above the context.coverage CheckInfo opened "Deliberately cites no rule" three lines above rule="OOLD-SCH-21d7".
• test_pipeline.py - the same claim in the docstring of test_missing_context_term_warns_and_names_the_orphan_property.

Why it matters

#120 (comment) required that the finding "must not fire at all when @vocab is declared". That was confirmed by an ad-hoc probe before the split of context.predicates into two checks, and never protected afterwards: no fixture in the corpus declared a @vocab at all, and the only @vocab in the suite asserted that the message text mentions it, which is a different claim. The permitted case was the one behaviour in that issue resting on nothing.

No production behaviour changes. The only non-test edit is a comment.

Verification

• full suite: 560 passed, 9 skipped
• make check: clean
• uv run oold validate tests/data/oold --offline: PASS, 380 ok

Parity

Green: 6 passed against a fresh clone of oold-schema main.

An earlier revision of this description claimed parity was broken on main. That was wrong and is retracted. Three parity tests did fail here, but the cause was a local ~/.cache/oold/meta/remote-main that was 15 days old and still declared x-oold-reverse-default-properties, removed upstream in OO-LD/oold-schema#152. oold meta fetch --force turned all three green with no change to this repository. main was never broken and CI was never at risk, since a fresh runner has no cache.

That the stale cache is silent is a real defect and is now #138. It is not this PR's to fix.

Refs #120. Does not close it: the severity question raised in #120 (comment) is still open.

[github-actions]

Release preview

No version bump from the current commits (stays at v0.18.1). Use conventional commit types (feat, fix, ...) to trigger a release.

Changelog preview (truncated)

_{Preview via python-semantic-release and conventional commits.}

[github-actions]

📊 Benchmark Results

Click to see benchmark comparison

📊 Benchmark Comparison (threshold: 1.3x)
============================================================

➖ Unchanged (within threshold):
  ➖ test_simple_dict_document_store: 0.0017s → 0.0017s (+0.8%)
  ➖ test_sqlite_document_store: 0.0019s → 0.0019s (+0.4%)
  ➖ test_local_sparql_store: 0.0388s → 0.0384s (-0.9%)
  ➖ test_oneof_subschema: 0.0602s → 0.0597s (-0.9%)
  ➖ test_enum_docstrings: 0.0528s → 0.0502s (-4.9%)
  ➖ test_subclass_inheritance: 0.0539s → 0.0535s (-0.8%)
  ➖ test_class_hierarchy: 0.0516s → 0.0518s (+0.5%)
  ➖ test_core[v1]: 0.0392s → 0.0402s (+2.5%)
  ➖ test_core[v2]: 0.0445s → 0.0450s (+1.3%)
  ➖ test_schema_generation[v1]: 0.0017s → 0.0017s (-0.7%)
  ➖ test_schema_generation[v2]: 0.0028s → 0.0028s (+0.2%)
  ➖ test_simple_json: 0.0007s → 0.0007s (-1.8%)
  ➖ test_complex_graph: 0.0016s → 0.0016s (-0.9%)

============================================================
Summary: 0 regressions, 0 improvements, 13 unchanged
============================================================

✅ No significant performance regressions

Threshold: 1.3x (30% slower triggers a regression warning)

Note: Benchmarks are informational only and won't fail the build.

_{💡 Tip: Download the benchmark-results artifact for detailed JSON data}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!