Establish a Ruff baseline and CI adoption policy

[bigsmartben]

Problem

The repository CI runs uvx ruff check src/. On PR #65, this reports 572 findings in existing src/ code, including files and lines outside the PR diff. The functional, integration, security, shell, and cross-platform test checks pass, but the repository-wide lint baseline prevents a clean merge signal.

This is a quality-governance concern, not a behavior-testability or requirement-gate change.

Goal

Establish a documented Ruff baseline and CI adoption policy that prevents new lint debt while allowing independently validated feature PRs to remain attributable to their own changes.

Scope

• Reproduce and record the current ruff check src/ baseline on main.
• Decide and document the transition model (for example: a tracked baseline/ratchet, scoped changed-file enforcement, or a staged per-rule/per-directory cleanup).
• Ensure CI fails when a PR introduces or worsens Ruff findings in the enforced scope.
• Define an incremental cleanup plan to converge the repository to a fully clean ruff check src/.
• Keep all existing Ruff rules visible; do not silently disable rules merely to make CI green.
• Update CI/documentation and add regression coverage for the chosen policy.

Non-goals

• Bulk-reformatting unrelated application code as part of Issue [Feature]: Separate requirement gates from plan-stage behavior testability #61.
• Changing the requirement-gate, BDD Plan, or workflow-preset behavior.

Acceptance criteria

• The baseline count and command are reproducible from the default branch.
• CI distinguishes pre-existing findings from PR-introduced findings.
• A deliberately introduced violation in an enforced file makes CI fail.
• An unrelated feature PR with no new violations can pass CI while historical findings remain tracked.
• The chosen baseline artifact/process has an owner and a removal/convergence path.
• The policy and contributor-facing commands are documented.

Related work

• Issue #61: feature delivery that exposed the baseline.
• PR #65: remains Draft pending this policy decision and review.

Created on behalf of @bigsmartben by Codex (model: GPT-5).