Split and refactor `Scenario` #102

	name: Benchmark
	on:
	push:
	branches:
	- main
	pull_request:
	paths:
	- crates/**
	- Cargo.toml
	- Cargo.lock
	- .github/workflows/bench.yml
	- pixi.lock
	permissions:
	contents: write
	deployments: write

	jobs:
	benchmark:
	name: Run Rust benchmark with criterion
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v5

	- uses: prefix-dev/[email protected]
	with:
	pixi-version: v0.50.2
	cache: true
	frozen: true
	cache-write: ${{ github.ref == 'refs/heads/main' }}
	environments: dev

	- uses: Swatinem/rust-cache@v2
	with:
	shared-key: "gha"
	save-if: false

	- name: Run benchmark
	run: \|
	pixi r -e dev cargo --locked install cargo-criterion
	pixi r -e dev cargo --locked criterion --output-format bencher --benches 2>&1 \| tee output.txt

	- name: Store benchmark result
	uses: benchmark-action/github-action-benchmark@v1
	if: ${{ github.ref == 'refs/heads/main' && github.event_name == 'push' }}
	with:
	name: Rust Benchmark
	tool: 'cargo'
	output-file-path: output.txt
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: true
	# Show alert with commit comment on detecting possible performance
	# regression
	# Benchmarking with GA is noisy, so only alert significant
	# regressions.
	# TODO: This elevated threshold is temporary (see PR for details).
	# Lower this value once the cache refactor restores performance.
	alert-threshold: '250%'
	comment-on-alert: true
	fail-on-alert: true
	alert-comment-cc-users: '@castelao @ppinchuk'

Provide feedback