feat: comprehensive API benchmark suite with autocannon (closes #30)

[qiqi1616]

/claim #30\n\n## Summary\n\nComprehensive API benchmark suite covering all platform endpoints using autocannon.\n\n### Benchmark Environment\n\nHardware\n- CPU: Cloud VM\n- RAM: 4GB\n- Storage: SSD\n- OS: Linux\n\nRuntime\n- Node.js 20+\n- No resource limits applied\n\nAI Agent\n- Agent: Hermes Agent\n- Model: deepseek-v4-pro\n- Provider: DeepSeek\n- Execution: human-supervised\n- Shell access: yes\n- Internet access: yes\n\n### What's Included\n\n- benchmarks/run-benchmark.mjs — Main benchmark runner using autocannon\n- benchmarks/endpoints.mjs — All 16 /api/ endpoints with realistic payloads\n- benchmarks/thresholds.json — Per-endpoint p99 latency thresholds\n- benchmarks/check-thresholds.mjs — CI threshold gate script\n- .env.benchmark.example — Configuration template\n- .github/workflows/benchmark-smoke.yml — CI smoke benchmark on PR\n- benchmarks/README.md — Documentation\n\n### Metrics Captured\n- p50, p95, p99 latency (ms)\n- Requests per second (RPS)\n- Error rate (%)\n- Time to first byte (TTFB)\n\n### Quick Start\nbash\nSMOKE=true node benchmarks/run-benchmark.mjs # CI smoke test\nnode benchmarks/run-benchmark.mjs # Full benchmark\n\n\n### Output\nResults written to benchmarks/results/summary.json and summary.md\n\nCloses #30

[BossChaos]

Code Review

PR: feat: comprehensive API benchmark suite with autocannon (closes #30) by @qiqi1616

• ✅ Performance benchmark suite

Wallet: 0xdaE5d307339074A24F579dB48e7c639359D94904

Code review for SecureBananaLabs bug-bounty