ci: try windows perf-scale nuts differences #358

	# runs some very-large-repo tests (for windows filesystem limits) and commits perf results for comparison
	name: perf-scale-nuts
	on:
	push:
	branches-ignore: [main]
	workflow_dispatch:

	# linux will finish 10 min ahead of windows, but prevent other branches/commits from hitting simultaneously
	# since we're pushing git commits and there would be conflicts
	concurrency: perf-scale-nuts

	jobs:
	perf-scale-nuts:
	strategy:
	matrix:
	# os: [ubuntu-latest, windows-latest]
	os: [windows-latest]
	fail-fast: false
	runs-on: ${{ matrix.os }}
	steps:
	- uses: actions/checkout@v3
	- uses: actions/setup-node@v3
	with:
	node-version: lts/*
	cache: yarn
	- uses: nick-fields/retry@943e742917ac94714d2f408a0e8320f2d1fcafcd
	name: add CLI as global dependency
	with:
	max_attempts: ${{ inputs.retries }}
	command: npm install @salesforce/cli@nightly -g
	timeout_minutes: 60
	- uses: salesforcecli/github-workflows/.github/actions/yarnInstallWithRetries@main
	- run: yarn test:nuts:scale

	# Run `github-action-benchmark` action
	- name: Store benchmark result
	uses: benchmark-action/github-action-benchmark@5bbce78ef18edf5b96cb2d23e8d240b485f9dc4a
	with:
	tool: 'customSmallerIsBetter'
	output-file-path: test/nuts/scale/output.json
	comment-on-alert: true
	fail-on-alert: true
	# Push and deploy GitHub pages branch automatically
	# this has a bug where it creates duplicate commits when summary-always and aut-push are both true
	# summary-always: true
	comment-always: true
	benchmark-data-dir-path: perf-${{ runner.os}}
	auto-push: true
	github-token: ${{ secrets.SVC_CLI_BOT_GITHUB_TOKEN }}

Provide feedback