Set up criterion for benchmarking

[friendlymatthew]

This commit uses criterion for its benchmarking suite. Also contains a README about benchmarking instructions.

[hyperlint-ai]

PR Change Summary

Set up criterion for benchmarking and implemented std::error::Error for jiter errors.

• Migrated to criterion for the benchmarking suite
• Implemented std::error::Error for jiter error types
• Added a README for benchmark instructions

Added Files

• crates/jiter/benches/README.md

---

How can I customize these reviews?

Check out the Hyperlint AI Reviewer docs for more information on how to customize the review.

If you just want to ignore it on this PR, you can add the hyperlint-ignore label to the PR. Future changes won't trigger a Hyperlint review.

Note specifically for link checks, we only check the first 30 links in a file and we cache the results for several hours (for instance, if you just added a page, you might experience this). Our recommendation is to add hyperlint-ignore to the PR to ignore the link check for this PR.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

📢 Thoughts on this report? Let us know!

[codspeed-hq]

CodSpeed Performance Report

Merging #197 will create unknown performance changes

_{Comparing criterion (a10651b) with main (961d830)}

Summary

🆕 70 new benchmarks
⁉️ 70 dropped benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
🆕	big_jiter_iter	N/A	36.9 ms	N/A
🆕	big_jiter_skip	N/A	19.2 ms	N/A
🆕	big_jiter_value	N/A	49.9 ms	N/A
🆕	big_serde_value	N/A	204.3 ms	N/A
🆕	bigints_array_jiter_iter	N/A	125.3 µs	N/A
🆕	bigints_array_jiter_skip	N/A	94.1 µs	N/A
🆕	bigints_array_jiter_value	N/A	208.9 µs	N/A
🆕	bigints_array_serde_value	N/A	719.3 µs	N/A
🆕	floats_array_jiter_iter	N/A	184.7 µs	N/A
🆕	floats_array_jiter_skip	N/A	106.2 µs	N/A
🆕	floats_array_jiter_value	N/A	308.3 µs	N/A
🆕	floats_array_serde_value	N/A	1.1 ms	N/A
🆕	massive_ints_array_jiter_iter	N/A	573 µs	N/A
🆕	massive_ints_array_jiter_skip	N/A	223.4 µs	N/A
🆕	massive_ints_array_jiter_value	N/A	764.8 µs	N/A
🆕	massive_ints_array_serde_value	N/A	2.2 ms	N/A
🆕	medium_response_jiter_skip	N/A	16.9 µs	N/A
🆕	medium_response_jiter_value	N/A	37.8 µs	N/A
🆕	medium_response_jiter_value_owned	N/A	50.7 µs	N/A
🆕	medium_response_serde_value	N/A	84.3 µs	N/A
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

[davidhewitt]

Thanks, overall looks good though we need to fix the codspeed measurement in CI.

Also, maybe split the error impls into their own PR for sake of making those changes more visible in their own right?

[davidhewitt]

We should use codspeed_criterion_compat here so that the benchmarks continue to be measured in CI.

[davidhewitt]

Looks like this introduces a dependency on half that is not MSRV compatible. It might be possible to tune things to make that not the case.

[davidhewitt]

Super, thanks! 👍