feat(bench): add Playwright-based benchmark harness and migrate scenarios from fast-element

[radium-v]

Pull Request

📖 Description

Introduces a new benchmark package with a Playwright + Chrome DevTools trace-based benchmarking harness, replacing the previous @tensile-perf/web-components setup embedded in fast-element.

Benchmark scenarios (basic, observable, render, repeat-reverse, repeat-shift, when-basic, when-conditional, when-switch) have been migrated from inline .bench.ts files in fast-element to standalone suites under the new packages/bench package. Each suite runs in a real browser via Playwright, collects trace data through CDP, and reports timing stats (min/median/mean/p95/max) for scripting, layout, style recalc, and paint.

Also adds @types/parse5 as a dev dependency for fast-ssr to fix the build.

👩‍💻 Reviewer Notes

• The bench package is private: true and not published — it's a dev-only tool for local and CI performance testing.
• Benchmarks can be run against the Vite dev server (npm run test) or a production build (BENCH_DIST=true npm run test).
• The server/ subdirectory is a self-contained static file server for serving the Vite build output.

📑 Test Plan

• Run npm run test -w @microsoft/fast-bench (or cd packages/bench && npm run test) to execute the benchmark suite locally and confirm output tables print correctly.
• Run npm run build -w @microsoft/fast-ssr to verify the @types/parse5 fix resolves the build error.
• All existing tests should continue to pass.

✅ Checklist

General

• I have included a change request file using $ npm run change
• I have added tests for my changes.
• I have tested my changes.
• I have updated the project documentation to reflect my changes.
• I have read the CONTRIBUTING documentation and followed the standards for this project.

⏭ Next Steps

• Add a CI pipeline step to run benchmarks on PRs for regression detection.
• Investigate baseline recording and comparison for automated performance regression alerts.