ENH: Add empty .pre-commit-config.yaml for release-5.4

[blowekamp]

Prevents pre-commit errors when backporting commits from branches where pre-commit hooks are configured.

PR Checklist

• No API changes were made (or the changes have been approved)
• No major design changes were made (or the changes have been approved)
• Added test (or behavior not changed)
• Updated API documentation (or API not changed)
• Added license to new files (if any)
• Added Python wrapping to new files (if any) as described in ITK Software Guide Section 9.5
• Added ITK examples for all new major features (if any)

Refer to the ITK Software Guide for
further development details if necessary.

[dzenanz]

Looks good to me. Someone else should preferably review.

[greptile-apps]

Greptile Summary

Summary

• Adds an empty pre-commit configuration for the release-5.4 branch.
• The configuration is accepted by pre-commit and completes an all-files run without executing hooks, preserving the intended branch behavior.

Merge safety

Safe to merge.

Confidence Score: 5/5

Safe to merge: the added configuration is valid and retains the intended no-hook behavior.

No issues were found. The configuration was validated and an all-files pre-commit run completed successfully without hook output.

Files Needing Attention: None.

T-Rex Logs

What T-Rex did

• I authored the executable precommit-empty-repos-validation script to perform the empty-repos validation.
• I captured before and after logs for the empty-repos validation to illustrate the state before and after execution.
• I validated that the added configuration declares only an empty repos list and compared it against the missing-configuration case.
• I ran pre-commit across all files and confirmed that it completed with no hooks running.

_{Ran code and verified through T-Rex}

_{Reviews (1): Last reviewed commit: "ENH: Add empty .pre-commit-config.yaml f..." | Re-trigger Greptile}

[hjmjohnson]

Love it! This is 95% of what I think should be done.

The last 5% is completely optional, and should not hold this up. I would want to backport the items that are "free" to backport.

[hjmjohnson]

Love it! This is 95% of what I think should be done.

The last 5% is completely optional, and should not hold this up. I would want to backport the items that are "free" to backport.

I dropped main's pre-commit-hooks block verbatim into a release-5.4 worktree and ran it against the whole tree. 17 of the 21 hooks pass clean today — they'd be pure no-ops on this branch, at zero cost and with byte-identical configuration to main. The other 4 fail, but the total debt is one URL, one file mode, and whitespace.

Full hook-by-hook result on release-5.4

Hook	Result
check-added-large-files, check-ast, check-case-conflict, check-json, check-merge-conflict, check-toml, check-xml, check-yaml, debug-statements, destroyed-symlinks, detect-private-key, mixed-line-ending, no-commit-to-branch	Passed
check-illegal-windows-names, forbid-new-submodules, forbid-submodules, name-tests-test	Skipped (no matching files)
check-vcs-permalinks	Failed — 1 line
check-shebang-scripts-are-executable	Failed — 1 file
end-of-file-fixer + trailing-whitespace	Failed — 17 files, +40/-52

Reproduce with:

git worktree add /tmp/r54 upstream/release-5.4
# copy the pre-commit-hooks repo block from main's .pre-commit-config.yaml
pre-commit run --all-files

The 4 failures, in full

1. check-vcs-permalinks — one non-permanent GitHub link:
   CMake/ITKSetStandardCompilerFlags.cmake:232 uses blob/master/... and wants a SHA permalink.

2. check-shebang-scripts-are-executable — one file:
   Utilities/Maintenance/VCL_ModernizeNaming.py has a shebang but is not marked executable. Fixed with git add --chmod=+x.

3. end-of-file-fixer + trailing-whitespace — 17 files, whitespace only, no compiled content:

   • 7 × Documentation/Doxygen/*.dox
   • 4 × Documentation/docs/releases/*.md
   • 2 × Modules/Nonunit/IntegratedTest/test/*.mm
   • Modules/Nonunit/IntegratedTest/wrapping/test/CMakeLists.txt
   • Modules/Numerics/FEM/test/README
   • Utilities/Debugger/ITK.natvis
   • Documentation/docs/_static/icon/html_head_content.txt

Why the whitespace hooks in particular are worth having here

This PR's own justification is that commits get backported from main. main enforces trailing-whitespace and end-of-file-fixer; release-5.4 currently does not. So a cherry-pick carries main-normalized whitespace into a branch holding 17 files of un-normalized whitespace, and the divergence accumulates precisely along the paths backports travel.

Worth noting the 4 failures are all genuine defects of the kind these hooks exist to catch — a master-branch link that will rot, a script that cannot be executed, and trailing whitespace. That 17 hooks pass untouched says release-5.4 is in better shape than its lack of enforcement would predict.

Tiny nit, take it or leave it: COMP: may fit better than ENH: here, since this is build/CI tooling rather than new functionality.

Either way — none of this should block merging as-is. The empty config solves the stated problem, and the hook enablement is a clean follow-up whenever it suits.

[blowekamp]

@hjmjohnson Your suggested followup would be nice.