build(deps): bump argminmax from 0.6.3 to 0.7.0 in /flexviz_polars in the cargo group

[dependabot]

Bumps the cargo group in /flexviz_polars with 1 update: argminmax.

Updates argminmax from 0.6.3 to 0.7.0

Release notes

Sourced from argminmax's releases.

v0.7.0

Highlights

• Skip nulls: the new ArgMinMaxMasked and NaNArgMinMaxMasked traits take an Arrow validity bitmap and its bit offset, and return None when no element is valid (#99). The arrow implementations use them, so they now skip the nulls.
• AVX512 on stable Rust (#83). The default features build on stable Rust for all targets.
• i128 and u128 support (#92).
• Faster: the SIMD loops compare the running min/max with one candidate per group of 4 vectors (#88). NEON is used for i64 and u64 on aarch64 (#104).
• Fixes: -0.0 and 0.0 are equal (f16, and f32/f64 with NaNs) (#98, #100); negative NaNs are ignored in f16 SIMD (#85); the ignore-NaN functions return the first non-NaN index when only infinities are left (#97, #102); 8 and 16-bit data only go to AVX512 when AVX512BW is available (#84).

Breaking changes

• MSRV is 1.89 (#83).
• nightly_simd is no longer a default feature (#91). It only enables NEON on 32-bit ARM (nightly Rust). On x86 and aarch64 it has no effect.
• The arrow2 feature is removed (#101), as arrow2 is archived. Call the functions on array.values().as_slice(), with ArgMinMaxMasked to skip the nulls.
• The arrow implementations skip nulls (#99). Before, the values in the null slots were included in the result.

What's Changed

• chore: update dependencies by @​jvdd in jvdd/argminmax#78
• ci: upgrade CodSpeed to v5 and pin nightly toolchain by @​jvdd in jvdd/argminmax#80
• ci: enable dependabot version updates by @​jvdd in jvdd/argminmax#79
• test: exclude infinities from the random float test data by @​jvdd in jvdd/argminmax#86
• chore(deps): bump the actions group with 2 updates by @​dependabot[bot] in jvdd/argminmax#81
• chore(deps): update rstest requirement from 0.26 to 0.27 in the cargo group by @​dependabot[bot] in jvdd/argminmax#82
• fix: ignore negative NaNs in the f16 SIMD implementations by @​jvdd in jvdd/argminmax#85
• fix: only dispatch 8 and 16-bit data to AVX512 when AVX512BW is available by @​jvdd in jvdd/argminmax#84
• ci: run the tests on AVX512 CPUs emulated by Intel SDE by @​jvdd in jvdd/argminmax#87
• feat: ungate AVX512 from nightly_simd and set rust-version to 1.89 by @​jvdd in jvdd/argminmax#83
• refactor: derive the core argmin/argmax from the core argminmax by @​jvdd in jvdd/argminmax#93
• ci: enable clippy by @​jvdd in jvdd/argminmax#90
• feat!: make nightly_simd a non-default feature by @​jvdd in jvdd/argminmax#91
• feat: add i128 and u128 support by @​jvdd in jvdd/argminmax#92
• refactor: dispatch all argminmax methods with one macro by @​jvdd in jvdd/argminmax#94
• fix: return the first non-NaN index in the scalar f16 argmin and argmax when only infinities are left by @​jvdd in jvdd/argminmax#97
• fix: treat -0.0 and 0.0 as equal in the f32 and f64 SIMD implementations that return NaNs by @​jvdd in jvdd/argminmax#98
• feat!: remove the arrow2 feature, as arrow2 is archived by @​jvdd in jvdd/argminmax#101
• fix: treat -0.0 and 0.0 as equal in the f16 implementations by @​jvdd in jvdd/argminmax#100
• fix: return the first non-NaN index in the SIMD implementations that ignore NaNs when only infinities are left by @​jvdd in jvdd/argminmax#102
• perf: compare the running min/max with one candidate per group of 4 vectors by @​jvdd in jvdd/argminmax#88
• perf: use NEON for the i64 and u64 argminmax on aarch64 by @​jvdd in jvdd/argminmax#104
• feat!: skip nulls with ArgMinMaxMasked and NaNArgMinMaxMasked, also in the arrow implementations by @​jvdd in jvdd/argminmax#99
• test: cover arrow Float16 arrays with nulls, and small cleanups of the masked code by @​jvdd in jvdd/argminmax#111
• refactor: implement DTypeInfo with one blanket impl instead of a macro by @​jvdd in jvdd/argminmax#112
• test: compare SIMD with scalar for every short array length, also on data with many ties by @​jvdd in jvdd/argminmax#114

New Contributors

• @​dependabot[bot] made their first contribution in jvdd/argminmax#81

Full Changelog: jvdd/argminmax@v0.6.3...v0.7.0

Commits

• dce8e65 chore: release v0.7.0
• cef791c test: compare SIMD with scalar for every short array length, also on data wit...
• f138adf refactor: implement DTypeInfo with one blanket impl instead of a macro (#112)
• d2c2cc0 test: cover arrow Float16 arrays with nulls, and small cleanups of the masked...
• e7e2daf feat!: skip nulls with ArgMinMaxMasked and NaNArgMinMaxMasked, also in the ar...
• 3ccf25d perf: use NEON for the i64 and u64 argminmax on aarch64 (#104)
• 69c720e perf: compare the running min/max with one candidate per group of 4 vectors (...
• 793bd92 fix: return the first non-NaN index in the SIMD implementations that ignore N...
• d812d3a fix: treat -0.0 and 0.0 as equal in the f16 implementations (#100)
• 414ea51 feat!: remove the arrow2 feature, as arrow2 is archived (#101)
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions

[codspeed]

Merging this PR will not alter performance

✅ 34 untouched benchmarks
⏩ 1 skipped benchmark¹

---

_{Comparing dependabot/cargo/flexviz_polars/cargo-01d41200a8 (16d60ae) with main (b8fe166)}

Footnotes

1. 1 benchmark was skipped, so the baseline result was used instead. If it was deleted from the codebase, click here and archive it to remove it from the performance reports. ↩