test(syntax): isolate historical measurement evidence

[zoeyrose]

Summary

• keep the committed syntax measurement report and captured representative-map metadata unchanged
• validate report provenance, measured byte ratios, section identities, and summary records internally without consulting live map bytes
• retain deterministic live representative selection as a separate test
• replace the corpus-state-dependent regression with a temporary-corpus header edit and fail-closed guards that reject live selection or source-byte reads during historical validation

Coordinates

• Base: main at ffc24fe7778154c31aa45e10b750dd5b304b98bd
• Head: fix/210-historical-syntax-measurements at 4b549a6ae712e55005b5f1c47622a4bae66b9925
• Worktree: /workspaces/atrinik/workspace/worktrees/content/issue-210-historical-syntax-measurements
• Commits: 48786edac test(syntax): isolate historical measurement evidence; 3ed46eb4b test(syntax): validate captures independently; 4b549a6ae test(syntax): guard historical validation boundary

Validation

• complete focused syntax-evaluation suite (17 tests passed)
• python3 tools/validate.py (167 tests plus complete contract, schema, syntax, catalog, lossless-core, provenance/license, and isolated runtime-collection gates)
• git diff --check
• confirmed prototypes/authored-syntax-v1/measurement-baseline.json has the same Git blob as the fetched base

Review

• complete raw base-to-head review across requirements, correctness, compatibility, maintainability, tests, lifecycle, performance, safety, and operations
• fixed historical/live corpus coupling, a tautological regression, live implementation-digest coupling, incomplete profile/state provenance checks, and the missing anti-recoupling guard
• two final independent fresh-context whole-diff reviews at 4b549a6ae: zero known actionable findings
• mutation probes confirmed the regression fails if historical validation consults live selection or source bytes

Verification

Runtime gameplay is not applicable because this changes only repository validation tests and no authored content or runtime behavior. From the issue worktree, run:

python3 -m unittest -v \
  tools.tests.test_syntax_evaluation.SyntaxEvaluationTest.test_representative_map_selection_is_deterministic_and_size_ordered \
  tools.tests.test_syntax_evaluation.SyntaxEvaluationTest.test_committed_measurement_baseline_preserves_captured_inputs \
  tools.tests.test_syntax_evaluation.SyntaxEvaluationTest.test_unrelated_map_header_edit_does_not_relabel_historical_measurements
python3 tools/validate.py
git diff --check

The commands are safe to repeat. They provision no profile, scenario, state, topology, server/client data, or credentials; no shutdown is required. Preserve the issue worktree and ignored review report while this PR remains open. Post-merge cleanup requires a separate preview-first request.

Closes #210

[github-actions]

🎉 This PR is included in version 4.2.2 🎉

The release is available on:

• v4.2.2
• GitHub release

Your semantic-release bot 📦🚀