fix(perf): correct metrics_recorded count and close SQLite handle on error in _store_metric

[groupthinking]

Canonical issue

Closes #1352

Outcome

POST /api/v1/performance/report now returns the number of samples actually persisted (len(samples)) instead of the number of metric keys submitted, so reports mixing numeric and non-numeric values no longer overcount. The legacy _store_metric now delegates to the batched _store_metrics with a one-element list, inheriting its try/finally so an exception mid-write no longer leaks the SQLite connection.

Scope

• Included: the two defects tracked in the linked issue, plus a regression test for the mixed-report count.
• Explicitly excluded: any behavior changes to the batched write path from perf: batch performance-metric writes into one SQLite round-trip #1341.

Risk

• Risk level: low
• Failure mode: clients relying on the old (incorrect) submitted-key count would see a lower metrics_recorded value for mixed reports.
• Rollback: revert the single commit.

Verification

At head cbb020f1ef3555a8e842d572651b4c00e347ee68:

• Focused tests — pytest tests/unit/test_v1_router_extended.py -k performance: 7 passed, including the new test_performance_report_counts_only_persisted_samples.
• Smoke test — _store_metric exercised against a temp SQLite DB; the row is written through the batched path.
• Required CI
• Review threads resolved

Production evidence

Not applicable — backend-only bug fix, covered by unit tests.

Agent handoff

• One canonical issue is linked
• No competing PR implements the same issue
• Acceptance criteria are satisfied
• Required checks pass on the current head
• Human decision is requested only for product, security, irreversible infrastructure, or production approval

Agent provenance

[linear-code]

GRV-308

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
v0-uvai	Ready	Preview, v0	Aug 29, 2026 7:54am

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 782507e9-8f44-45dc-955d-b4de88649a98

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Snapshot Warnings

⚠️: No snapshots were found for the head SHA dda8fd5.

Ensure that dependencies are being submitted on PR branches and consider enabling retry-on-snapshot-warnings. See the documentation for more information and troubleshooting advice.

Scanned Files

None

[github-actions]

Agent Completion Truth Gate: BLOCKED

Reasons: invalid_payload

Machine-readable verdict

{
  "details": {
    "collection_errors": [
      "linked_issue_not_agent_task",
      "missing_intent_snapshot",
      "missing_agent_run_id",
      "missing_agent_login"
    ],
    "invalid_fields": [
      "policy.agent_login",
      "policy.run_id"
    ]
  },
  "reasons": [
    "invalid_payload"
  ],
  "verdict": "blocked"
}

Workflow evidence