feat: add cairo-metrics benchmark harness + CI tracking

[giladchase]

Summary

• Introduce cairo-metrics: a small CLI that currently checks wall-clock regressions from main.
• Persist results in repo_root/results.db (git-ignored), currently SQLite keyed by run
  id (defaults to current git SHA). This enables caching and tracking results across time locally by
  checking out builds starting from this commit, and having the tool populate the db.
  The tool then allows comparisons based on the db results. Future work can easily extend this to
  a stateful DB machine (like RDS), since the DB is behind a trait.
• Add a GitHub Actions workflow that benchmarks baseline vs PR, reuses cached baseline results
  via artifacts (saves ci runtime for multiple PR runs over the same base branch), and posts a
  “Benchmark Comparison” PR comment, not blocking, the reviewer decided.
• Seed initial benchmark suites (corelib + OpenZeppelin): corelib uses local src, and openzepplin
  is bundeled as a vendored release (not a submodule for simplicity).
• Walltime engine is either home-brewed timed comparison of the compiler library,
  or via hyperfine which uses the binary. It uses hyperfine by
  default if available (need to install with apt) otherwise the builtin.
  This is useful locally, to debug the builtin engine itself
  (results are similar since hyperfine cancels out shell overhead), and since
  hyperfine outputs useful statistical anomaly messages.
  But if hyperfine is a pain to maintain it can be removed.

---

Type of change

Please check one:

• Bug fix (fixes incorrect behavior)
• New feature
• Performance improvement
• Documentation change with concrete technical impact
• Style, wording, formatting, or typo-only change

⚠️ Note:
To keep maintainer workload sustainable, we generally do not accept PRs that
are only minor wording, grammar, formatting, or style changes.
Such PRs may be closed without detailed review.

---

Why is this change needed?

Implemented a benchmarking harness for testing and tracking performance regressions

---

What was the behavior or documentation before?

---

What is the behavior or documentation after?

---

Related issue or discussion (if any)

---

Additional context

[reviewable-StarkWare]

This change is 

[giladchase]

• feat(metrics): add compilation phases #9403
• feat(metrics): add incremental mode logic #9402
• feat: add cairo-metrics benchmark harness + CI tracking #9401 👈 (View in Graphite)
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[giladchase]

@giladchase made 1 comment.
Reviewable status: 0 of 24 files reviewed, all discussions resolved (waiting on @orizi and @TomerStarkware).

---

crates/bin/cairo-metrics/src/engine.rs line 63 at r1 (raw file):

            // TODO(gilad): Incremental compilation requires cairo compiler support.
            // For now, skip incremental scenarios entirely.
            if is_incremental {

Next level in the stack adds the logic itself, but the short circuit will stay until we support incremental in the compiler.