chore(#6228): add qodo pr-agent config to disable stale Go test rules

[fullsend-ai-coder]

Summary

Add a .pr_agent.toml repo-level configuration file to disable qodo-code-review convention pack rules TC-001 and TC-002, which mandate stdlib-only Go testing (no testify). The project uses testify/assert and testify/require in 72% of test files, making these rules a standing source of false-positive findings on Go test PRs.

Related Issue

Closes #6228

Changes

• Add .pr_agent.toml with [pr_reviewer] and [pr_code_suggestions] sections
• Both sections include extra_instructions telling qodo to skip TC-001/TC-002 enforcement and not suggest replacing testify with stdlib assertions
• Configuration follows qodo's repo-level config format, overriding only the relevant settings

Testing

• TOML syntax validated (Python tomllib)
• Secret scan passed
• Verify on next Go test PR: qodo-code-review should produce zero TC-001/TC-002 findings

Checklist

• PR title follows Conventional Commits (correct type, ! for breaking changes)
• Commits are signed off (DCO) — human and human-directed agent sessions only
• I wrote this contribution myself and can explain all changes in it

Note: pre-commit could not run in the sandbox (network restriction on git fetch for hook environment setup). The post-script runs pre-commit authoritatively on the runner.

---

Closes #6228

Post-script verification

• Branch is not main/master (agent/6228-disable-stale-qodo-rules)
• Secret scan passed (gitleaks — b5314d3b59172d150587d159903642319b28d1d1..HEAD)
• PR body secret scan passed (gitleaks — no-git)

[fullsend-ai-review]

🤖 Finished Review · ✅ Success · Started 4:07 PM UTC · Completed 4:18 PM UTC

Commit: f341790 · View workflow run →

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[fullsend-ai-review]

Looks good to me

---

Labels: PR adds tooling configuration for the qodo code review bot (CI pipeline component)

[fullsend-ai-retro]

🤖 Finished Retro · ✅ Success · Started 12:31 PM UTC · Completed 12:46 PM UTC

Commit: f341790 · View workflow run →

[fullsend-ai-retro]

Retro: PR #6230 — add qodo pr-agent config to disable stale Go test rules

This PR was a clean end-to-end success across the full agent pipeline with zero rework.

Timeline

1. Retro agent auto-filed issue #6228 from findings on PR #6147 (Aug 14, 15:39 UTC)
2. Triage agent ran in ~4 min, rated severity "low", confirmed no .pr_agent.toml existed, noted related issue Fullsend review agent and Qodo bot produce overlapping code-level findings, creating noise on PRs with both reviewers active #4107 (run 31815595418)
3. Human (ralphbean) triggered /fs-code 12 min later (15:56)
4. Code agent produced PR chore(#6228): add qodo pr-agent config to disable stale Go test rules #6230 with a single correct commit — valid TOML, correct qodo config format, well-documented PR body citing 72% testify usage across 156/216 test files (run 31816991502)
5. Review agent approved with "Looks good to me" in ~12 min (run 31817763665)
6. Human (ralphbean) approved without comments 7 days later (Aug 21, 12:29)
7. PR merged immediately after human approval

What went well

• Full pipeline worked: Retro → Issue → Triage → Code → Review executed autonomously with human triggering only the /fs-code step.
• First-attempt correctness: The code agent produced a valid, well-documented change on the first try — no review changes, no fix cycles.
• Triage quality: Correctly identified severity, confirmed preconditions, and noted the related issue Fullsend review agent and Qodo bot produce overlapping code-level findings, creating noise on PRs with both reviewers active #4107.

Existing issue evidence (no new proposals needed)

• Surface agent-approved PRs awaiting human review after N days #966 (surface stale agent-approved PRs): This PR sat 7 days between agent approval (Aug 14) and human merge (Aug 21) for a trivial config-only change. Strong evidence the surfacing mechanism proposed in Surface agent-approved PRs awaiting human review after N days #966 would help.
• Review orchestrator: add fast path for trivial docs/prompt-only changes #1900 (fast path for trivial changes): A single new 18-line TOML file with zero code impact is exactly the class of change that would benefit from a streamlined review path.
• Fullsend review agent and Qodo bot produce overlapping code-level findings, creating noise on PRs with both reviewers active #4107 (qodo/fullsend review overlap): This PR is itself a partial mitigation for Fullsend review agent and Qodo bot produce overlapping code-level findings, creating noise on PRs with both reviewers active #4107 — silencing qodo's false-positive rules reduces review noise on Go test PRs.