feat: add cairo-to-cache and cache-to-sierra compile benchmarks

[orizi]

TL;DR

Added two new benchmarks to measure the performance of the lowering cache pipeline for Cairo compilation.

What changed?

Two new benchmark functions were added to the compile benchmark suite:

• fib: cairo-to-cache — measures the time to compile Cairo source to a serialized lowering cache (bytes), without Sierra generation.
• fib: cache-to-sierra — measures the time to generate Sierra from a pre-compiled lowering cache, isolating the cost of Sierra generation from the full compilation pipeline.

How to test?

Run the compile benchmarks:

cargo bench --bench compile

The two new benchmark entries (fib: cairo-to-cache and fib: cache-to-sierra) will appear in the output alongside the existing benchmarks.

Why make this change?

The lowering cache splits the compilation pipeline into two distinct phases. These benchmarks allow measuring each phase independently, making it possible to quantify the performance impact of caching lowering results and to track regressions or improvements in each phase separately.

[reviewable-StarkWare]

This change is 

[orizi]

• feat: add corelib benchmarks and refactor to shared bench_all_phases helper #9859
• feat: add cache-to-testing compile benchmark #9858
• feat: add cairo-to-cache and cache-to-sierra compile benchmarks #9857 👈 (View in Graphite)
• feat: add cairo-to-testing benchmark #9856
• feat: add diagnostics phase benchmark #9855
• main

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

[cursor]

PR Summary

Low Risk
Low risk: changes are limited to benchmark code and add no production-path logic, though they exercise new lowering-cache wiring that could affect benchmark stability/perf comparisons if misconfigured.

Overview
Adds two new compile benchmarks to split the pipeline around the lowering cache: fib: cairo-to-cache measures generating/serializing the lowering cache bytes, and fib: cache-to-sierra measures Sierra generation when a precomputed cache is injected via crate config.

Updates the benchmark harness to use generate_crate_cache, virtual cache blobs, and compile_prepared_db_program so the new phases can be timed independently alongside existing compile/diagnostics/testing benches.

^{Reviewed by Cursor Bugbot for commit 2c56d92. Bugbot is set up for automated code reviews on this repo. Configure here.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2c56d92706

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[TomerStarkware]

@TomerStarkware reviewed 1 file and all commit messages, and made 1 comment.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on eytan-starkware and orizi).