Add Copilot agent to lint CI for automatic lint fix on PRs

[Copilot]

• Understand current lint.yml workflow structure
• Add a copilot-fix-lint job that triggers when lint-python-format fails on a PR
• Fix review comment 1: Use always() && needs.lint-python-format.result == 'failure' instead of failure() to gate explicitly on the lint job result
• Fix review comment 2: Add issues: write permission since the script uses github.rest.issues.* APIs
• Fix review comment 3: Use github.paginate instead of single listComments call to handle PRs with 100+ comments
• Fix review comment 4: Add concurrency group to prevent race condition between concurrent workflow runs
• Validate YAML syntax

[Copilot]

Pull request overview

Adds a follow-up GitHub Actions job that, when linting fails on same-repo pull requests, posts a deduplicated PR comment mentioning @copilot with steps to reproduce and fix lint issues.

Changes:

• Introduces a new copilot-fix-lint job in the lint workflow.
• Adds logic to avoid duplicate “request Copilot” comments via an HTML marker.
• Includes a link to the failed workflow run and local reproduction commands in the PR comment.

[Copilot]

The job-level condition is broader than the PR description and may not run when intended. failure() can become true due to failures in other jobs (for example optional-lint), and dependent jobs are often skipped when needs fails unless you include always(). Prefer gating explicitly on the dependency result, for example checking needs.lint-python-format.result == 'failure' and combining with always() plus the pull_request and non-fork checks.

Suggested change

	# Only run when lint fails on pull requests from the same repo (not forks)
	if: >-
	failure()
	# Only run when lint-python-format fails on pull requests from the same repo (not forks)
	if: >-
	always()
	&& needs.lint-python-format.result == 'failure'

[Copilot]

actions/github-script is calling github.rest.issues.* APIs (listComments/createComment), which require the issues permission. This job currently sets only pull-requests: write, and job-level permissions overrides other scopes to none, so the API calls can fail with 403. Add issues: write (and keep pull-requests: write only if needed).

Suggested change

	permissions:
	permissions:
	issues: write

[Copilot]

Deduplication can miss prior markers because only the first 100 comments are fetched. If a PR has more than 100 comments, the marker could be on a later page and this job would post a duplicate request. Consider paginating through comments (or searching most-recent pages first), or updating an existing comment instead of creating a new one.

Suggested change

	// Avoid duplicate comments on workflow re-runs
	const { data: comments } = await github.rest.issues.listComments({
	owner: context.repo.owner,
	repo: context.repo.repo,
	issue_number: prNumber,
	per_page: 100,
	});
	// Avoid duplicate comments on workflow re-runs by checking all comment pages
	const comments = await github.paginate(
	github.rest.issues.listComments,
	{
	owner: context.repo.owner,
	repo: context.repo.repo,
	issue_number: prNumber,
	per_page: 100,
	},
	);

[Copilot]

There is a race window where two workflow runs (for example a re-run started before the prior run posts the comment) can both pass the "no marker" check and create duplicate comments. If duplicates are a concern, consider adding a workflow concurrency group for this job/workflow or switching to an idempotent update strategy (for example editing an existing marker comment).