ci: move benchmark regression job out of default PR workflow

Extract the benchmark-regression job from the default PR CI workflow
into a dedicated scheduled workflow (.forgejo/workflows/benchmark-scheduled.yml).

The benchmark job was blocking PR feedback for 99-132 minutes even when
lint/typecheck/tests passed in under 20 minutes. The new workflow runs:
- Nightly regression tests (2 AM UTC) comparing HEAD against master
- Weekly full benchmark suite (3 AM UTC Sundays) for trend analysis
- Manual dispatch trigger for ad-hoc benchmark validation

AWS S3 integration stores benchmark results for historical trend analysis.
Documentation updated in docs/development/ci-cd.md.
CONTRIBUTORS.md updated with benchmark workflow separation contribution.

ISSUES CLOSED: #9040

# Conflicts:
#	CONTRIBUTORS.md

# Conflicts:
#	CONTRIBUTORS.md

docs/development/ci-cd.md

@@ -399,3 +399,74 @@ bash scripts/setup-dev.sh
 pre-commit install
 pre-commit install --hook-type commit-msg
 ```
+
+## Benchmark Workflow
+
+### Why Benchmarks Are Separate from PR Validation
+
+The `benchmark-regression` job was removed from the default PR workflow because:
+
+- **Long runtime**: ASV benchmark runs take 99-132 minutes, blocking PR feedback
+  even when lint/typecheck/tests pass in under 20 minutes.
+- **Runner contention**: Long-running ASV tasks occupy docker runners, exacerbating
+  queue times for other PRs.
+- **Low signal for most changes**: The vast majority of code changes do not affect
+  performance; running benchmarks on every PR is wasteful.
+
+The benchmark workflow now runs on a schedule and can be triggered manually for
+performance-sensitive changes.
+
+### Benchmark Workflow Triggers
+
+The benchmark workflow (`.forgejo/workflows/benchmark-scheduled.yml`) runs:
+
+| Trigger | Schedule | Job | Purpose |
+|---------|----------|-----|---------|
+| Nightly cron | 2 AM UTC daily | `benchmark-regression` | Compare HEAD against master for regressions |
+| Weekly cron | 3 AM UTC Sundays | `benchmark-full` | Full benchmark suite for trend analysis |
+| Manual dispatch | On demand | Either | Ad-hoc validation for performance-sensitive PRs |
+
+### How to Trigger Benchmarks Manually
+
+To run benchmarks for a specific PR or branch:
+
+1. Navigate to **Actions** > **Benchmark Regression** in the Forgejo UI.
+2. Click **Run workflow**.
+3. Select the branch to benchmark.
+4. Set `base_sha` to the SHA or branch to compare against (default: `master`).
+5. Set `run_full_suite` to `true` for the full suite, or `false` (default) for
+   regression-only comparison.
+
+Alternatively, use the Forgejo API:
+
+```bash
+curl -X POST \
+  -H "Authorization: token ${FORGEJO_TOKEN}" \
+  -H "Content-Type: application/json" \
+  -d '{"ref": "your-branch", "inputs": {"base_sha": "master", "run_full_suite": "false"}}' \
+  "${FORGEJO_URL}/api/v1/repos/cleveragents/cleveragents-core/actions/workflows/benchmark-scheduled.yml/dispatches"
+```
+
+### AWS Credentials Required for S3 Benchmark Storage
+
+Benchmark results are stored in S3 for historical trend analysis. The following
+secrets must be configured in **Repository Settings > Actions > Secrets**:
+
+| Secret | Purpose |
+|--------|---------|
+| `AWS_ACCESS_KEY_ID` | AWS credentials for S3 benchmark storage |
+| `AWS_SECRET_ACCESS_KEY` | AWS credentials for S3 benchmark storage |
+| `AWS_DEFAULT_REGION` | AWS region for S3 benchmark storage |
+| `ASV_S3_BUCKET` | S3 bucket name for ASV benchmark storage |
+
+If these secrets are not configured, benchmark runs will still execute locally
+but results will not be published to S3 (a warning is emitted instead of failing).
+
+### Benchmark Results
+
+Benchmark results are available as:
+
+- **Forgejo Actions artifacts**: Uploaded as `benchmark-regression-results` or
+  `benchmark-full-results` with 90-day retention.
+- **S3 storage**: Published to `s3://${ASV_S3_BUCKET}/asv/` when AWS credentials
+  are configured. The HTML report is at `s3://${ASV_S3_BUCKET}/asv/html/`.