[plan] Review and rebalance label distribution

[github-actions]

Objective

Address label skew where 124 issues each have plan and ai-generated labels, limiting the usefulness of these labels for filtering and triage.

Context

Current label distribution shows heavy concentration:

• plan: 124 issues
• ai-generated: 124 issues
• Other labels: distributed across remaining issues

When labels apply to >50% of issues, they lose filtering value.

Approach

1. Analyze label usage patterns:

   gh issue list --label plan --limit 200 --json number,title,labels
   gh issue list --label ai-generated --limit 200 --json number,title,labels

2. Review if plan and ai-generated labels are being applied too broadly:
   • Should these be temporary labels removed after processing?
   • Are there sub-categories that would be more useful?
   • Should some issues have these labels removed?
3. Propose label taxonomy improvements:
   • Consider plan-* subcategories (e.g., plan-active, plan-completed)
   • Consider retiring overused labels in favor of more specific ones
   • Suggest labels that provide better signal for triage
4. Document label usage guidelines in CONTRIBUTING.md

Files to Modify

• CONTRIBUTING.md - Add label usage guidelines
• GitHub repository settings (if creating/modifying labels)

Acceptance Criteria

• Label distribution analysis completed and documented
• Recommendations provided for label taxonomy improvements
• Label usage guidelines added to CONTRIBUTING.md
• If actionable changes identified, implementation plan provided
• No breaking changes to existing automation that depends on these labels

Estimated Effort

Quick (1-2 hours)
Related to #6857

AI generated by Plan Command for discussion #6855