chore: Isolate skill file normalization from layout handling

[hatayama]

Summary

• Skill file line-ending and encoding normalization now has a focused owner instead of living inside layout detection.
• Existing generated skill content remains byte-compatible.

User Impact

• There is no intended behavior change. Generated skill copies continue to normalize supported text files to LF while preserving UTF-16 endianness, BOMs, non-text files, and binary content.
• The smaller layout class makes future skill setup maintenance easier to review without mixing byte-encoding logic into directory layout rules.

Changes

• Added focused characterization coverage before moving production code.
• Extracted the text extension allowlist, UTF-16 detection, and byte-level normalization helpers into SkillFileContentNormalizer without changing their logic or names.
• Updated both production callers and the focused test fixture to reference the extracted class directly.
• Removed the old entry point from SkillInstallLayout instead of leaving a compatibility wrapper.
• Reduced SkillInstallLayout from 571 lines to 397 lines.

Verification

• Characterization tests were Green before and after extraction.
• Repo-local Unity compile: 0 errors, 0 warnings.
• SkillFileContentNormalizerTests: 6 passed.
• SkillInstallLayoutTests: 18 passed.
• ToolSkillSynchronizerTests: 52 passed.
• git diff --check.

Compatibility

• Source discovery behavior, generated byte content, and wire formats are unchanged, so no protocol version bump is required.

Deferred Review Finding

• cubic identified a pre-existing UTF-16BE ambiguity where the little-endian line-ending heuristic runs before the explicit BE BOM check. The same ordering existed before this move-only extraction, so it is recorded as R2-28 for a separate Red-first bug-fix PR rather than mixing behavior changes into this refactor.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR extracts skill file line-ending normalization logic into a new SkillFileContentNormalizer class, handling both single-byte and UTF-16 (LE/BE with BOM) text normalization. SkillInstallLayout is updated to delegate to this new class instead of its previous inline implementation, and tests are relocated/added accordingly.

Changes

Skill file content normalizer extraction

Layer / File(s)	Summary
Normalizer contract and gating logic Packages/src/Editor/Infrastructure/SkillSetup/SkillFileContentNormalizer.cs	New class defines UTF-16 BOM/encoding constants, a text extension allowlist, and NormalizeSkillFileContent entry point that routes normalization based on extension and content markers.
Single-byte and UTF-16 normalization implementation Packages/src/Editor/Infrastructure/SkillSetup/SkillFileContentNormalizer.cs	Implements CR/CRLF→LF conversion for single-byte text and UTF-16 code-unit-level normalization preserving BOM and endianness, plus byte read/write helpers.
SkillInstallLayout delegation Packages/src/Editor/Infrastructure/SkillSetup/SkillInstallLayout.cs	Removes inline UTF/BOM constants and text extension filter; installed and source skill file collection now call SkillFileContentNormalizer.NormalizeSkillFileContent.
Test coverage relocation Assets/Tests/Editor/SkillFileContentNormalizerTests.cs, Assets/Tests/Editor/SkillInstallLayoutTests.cs	Adds six new tests covering encodings, line endings, and gating rules for the normalizer; removes the redundant UTF-16LE test and unused using directive from the layout test file.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• hatayama/unity-cli-loop#1039: Both PRs implement/consume deterministic skill text line-ending normalization.
• hatayama/unity-cli-loop#1588: Both PRs relate to skill file content normalization (PowerShell UTF-16LE/line-ending handling) and its test verification in SkillInstallLayout.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title clearly summarizes the refactor: isolating skill file normalization from layout handling.
Description check	✅ Passed	The description is directly about the same extraction, behavior preservation, and test updates in this changeset.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch refactor/hatayama/extract-skill-content-normalization

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[cubic-dev-ai]

All reported issues were addressed across 6 files

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>