Bump maturin version: 0.14 -> 1.1

[TomaSajt]

This PR bumps the maturin version from 0.14 to 1.1

It also replaces the usage of [package.metadata.maturin].name with [tool.maturin].module-name.

(the former one has been deprecated since 0.15)

---

I made this change, as I'm trying to package this Python package for nixpkgs and it uses a newer version of maturin, so I had to patch the pyproject.toml with the output path. This PR just that patch as a commit.

[codspeed-hq]

CodSpeed Performance Report

Merging #52 will degrade performances by 76.06%

_{Comparing TomaSajt:main (de0be35) with main (33a7cfa)}

🎉 Hooray! pytest-codspeed just leveled up to 2.0.0!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🥳!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

Summary

🔥 1 improvements
❌ 15 regressions
✅ 374 untouched benchmarks

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

Benchmarks breakdown

	Benchmark	main	TomaSajt:main	Change
❌	test_minmax_no_x[False-float32-100-100,000]	794 µs	1,615.9 µs	-50.86%
❌	test_minmax_no_x[False-float64-100-100,000]	1.3 ms	1.9 ms	-30.15%
❌	test_minmax_no_x[False-int32-100-100,000]	1.5 ms	3.8 ms	-60.7%
❌	test_minmax_no_x[False-int64-100-100,000]	2 ms	2.5 ms	-20.45%
❌	test_minmax_no_x[True-float32-100-100,000]	882.2 µs	1,169.3 µs	-24.56%
❌	test_minmax_with_x[True-float64-1,000-100,000]	6.7 ms	8.3 ms	-19.13%
❌	test_minmax_with_x[True-int32-1,000-100,000]	6.9 ms	8.5 ms	-18.73%
❌	test_minmax_with_x[True-int64-1,000-100,000]	7.4 ms	9 ms	-17.25%
❌	test_m4_with_x[True-float32-1,000-100,000]	5.7 ms	7.2 ms	-20.67%
❌	test_m4_with_x[True-float64-1,000-100,000]	6.1 ms	7.9 ms	-22.71%
❌	test_m4_with_x[True-int32-1,000-100,000]	6.4 ms	8.1 ms	-20.81%
❌	test_m4_with_x[True-int64-1,000-100,000]	7 ms	8.7 ms	-19.22%
❌	test_minmaxlttb_with_x[True-float64-100-100,000]	16.6 ms	69.5 ms	-76.06%
❌	test_minmaxlttb_with_x[True-int32-100-100,000]	16.9 ms	61.9 ms	-72.73%
🔥	test_minmaxlttb_with_x[True-int32-1,000-100,000]	154.9 ms	123.3 ms	+25.62%
❌	test_minmaxlttb_with_x[True-int64-100-100,000]	17.4 ms	38.5 ms	-54.66%

[jvdd]

Do not really understand how this PR might impact performance? Do you have any idea @TomaSajt 🙃

[TomaSajt]

Do not really understand how this PR might impact performance? Do you have any idea @TomaSajt

I don't know why it's slower either.

Maybe there's some breaking change, which disables some optimization by default.

https://www.maturin.rs/changelog.html

One guess for the culprit is the new default maturin flag --no-default-features, which somehow (I don't know why it would affect dependencies) disables the nightly_simd feature on argminmax.

Or maybe we just need to use the 0.19 version of pyo3 and numpy,

I don't know, what the problem might be, so I'll make some commits, to look at the new benchmarks. If something doesn't help, I'll revert it.

[TomaSajt]

I realize now, that this this repo is still using argminmax 3.1, which doesn't have that flag yet, so I reverted that change. Maybe just changing those package versions will help (but I find it unlikely)

[jvdd]

Hey @TomaSajt,

I observe that the "regressions" are only for 100k values, and there are no regressions for 1M values benchmarks. I suppose this can be attributed to a measurement error rather than a real regression (I observed smth similar in #51).

If tests pass, this one is going in!

Thx for submitting this PR 🤝

P.S.: we plan to update the code base to the latest argminmax version #50 (which is the true power behind this library), this will add some more performance improvements + step away from the (kinda outdated) ndarray dependecy :))

[jvdd]

tada Hooray! pytest-codspeed just leveled up to 2.0.0!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability partying_face!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

Actually think it might be due to pytest-codspeed 2.0