feat: add command to compare multiple benchmarks

[nazarhussain]

Motivation

Add a new command to support comparing multiple benchmarks.

Description

This will helps us to compare benchmark results across different runtimes.

Steps to test or reproduce

Run all tests and workflows.

[github-actions]

Performance Report

✔️ no performance regression detected

Full benchmark results

Benchmark suite	Current: 7cac592	Previous: null	Ratio
sum array with raw for loop	933.79 us/op
sum array with reduce	10.186 ms/op
sum array with reduce beforeEach	10.219 ms/op
sum array with reduce before beforeEach	10.224 ms/op
sum array with reduce high threshold	10.276 ms/op
sum array with reduce no threshold	10.175 ms/op

by benchmarkbot/action

[github-actions]

Benchmark Comparison Report

Benchmark suite	nodejs-benchmark	deno-benchmark		bun-benchmark
	Average Ns	Average Ns	Ratio	Average Ns	Ratio
sum array with raw for loop	937.14 us/op	935.92 us/op	1.00	592.07 us/op	0.63
sum array with reduce	9.2522 ms/op	9.2838 ms/op	1.00	3.6311 ms/op	0.39
sum array with reduce beforeEach	9.2872 ms/op	9.3488 ms/op	1.01	3.6738 ms/op	0.40
sum array with reduce before beforeEach	9.3053 ms/op	9.3442 ms/op	1.00	3.6886 ms/op	0.40
sum array with reduce high threshold	9.2776 ms/op	9.2781 ms/op	1.00	7.1059 ms/op	0.77
sum array with reduce no threshold	9.2947 ms/op	9.3507 ms/op	1.01	7.0874 ms/op	0.76

by benchmarkbot/compare

[nazarhussain]

Created this workflow as reusable workflow so we can utilize it in other repos.

[matthewkeil]

This is pretty sweet. Do we need some docs on how to use it? Maybe a comment block at the head for consumers that goes through what it does and how to run it so they dont have to study the code to grok whats cooking

[matthewkeil]

LGTM 🎸 Only the couple spelling errors but approving and can fix them at your leisure