Add extra inlining to speed up take

[Dandandan]

They had some inline hints on the methods, but not all the way to extend_from_trusted_len_iter_unchecked and try_from_trusted_len_iter_unchecked.

This has changes around 30-50%

Benchmarking take i32 512: Collecting 100 samples in estimated 5.0009 s (17M ite                                                                                take i32 512            time:   [300.35 ns 301.92 ns 303.62 ns]
                        change: [-51.942% -51.683% -51.411%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe

Benchmarking take i32 1024: Collecting 100 samples in estimated 5.0032 s (7.4M i                                                                                take i32 1024           time:   [674.49 ns 675.96 ns 677.54 ns]
                        change: [-47.687% -47.558% -47.445%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  8 (8.00%) high mild

Benchmarking take i32 nulls 512: Collecting 100 samples in estimated 5.0013 s (9                                                                                take i32 nulls 512      time:   [511.26 ns 511.65 ns 512.11 ns]
                        change: [-1.1965% -0.9889% -0.7859%] (p = 0.00 < 0.05)
                        Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild

Benchmarking take i32 nulls 1024: Collecting 100 samples in estimated 5.0011 s (                                                                                take i32 nulls 1024     time:   [796.57 ns 797.00 ns 797.49 ns]
                        change: [-12.886% -12.768% -12.652%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  3 (3.00%) high mild
  8 (8.00%) high severe

Benchmarking take str 512: Collecting 100 samples in estimated 5.0041 s (1.6M it                                                                                take str 512            time:   [3.0675 us 3.0691 us 3.0705 us]
                        change: [-19.937% -19.801% -19.674%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild

Benchmarking take str 1024: Collecting 100 samples in estimated 5.0202 s (1.0M i                                                                                take str 1024           time:   [4.9393 us 4.9453 us 4.9512 us]
                        change: [-38.380% -38.271% -38.141%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) low mild
  1 (1.00%) high mild
  2 (2.00%) high severe

Benchmarking take str null indices 512: Collecting 100 samples in estimated 5.01                                                                                take str null indices 512                        
                        time:   [2.9738 us 2.9800 us 2.9856 us]
                        change: [-33.972% -33.766% -33.544%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) high mild
  3 (3.00%) high severe

Benchmarking take str null indices 1024: Collecting 100 samples in estimated 5.0                                                                                take str null indices 1024                        
                        time:   [5.2084 us 5.2177 us 5.2279 us]
                        change: [-35.800% -35.646% -35.513%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) low mild
  3 (3.00%) high mild

Benchmarking take str null values 1024: Collecting 100 samples in estimated 5.03                                                                                take str null values 1024                        
                        time:   [11.351 us 11.362 us 11.375 us]
                        change: [-32.987% -32.304% -31.477%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  3 (3.00%) high mild
  4 (4.00%) high severe

Benchmarking take str null values null indices 1024: Collecting 100 samples in e                                                                                take str null values null indices 1024                        
                        time:   [10.549 us 10.584 us 10.617 us]
                        change: [-22.214% -21.878% -21.539%] (p = 0.00 < 0.05)
                        Performance has improved.

[codecov]

Codecov Report

Merging #226 (558bfa2) into main (eaa9be9) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main     #226   +/-   ##
=======================================
  Coverage   76.94%   76.94%           
=======================================
  Files         229      229           
  Lines       19536    19536           
=======================================
  Hits        15031    15031           
  Misses       4505     4505           

Impacted Files	Coverage Δ
src/buffer/mutable.rs	91.76% <ø> (ø)
src/compute/take/generic_binary.rs	98.86% <ø> (ø)
src/compute/take/primitive.rs	93.87% <ø> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update eaa9be9...558bfa2. Read the comment docs.

[jorgecarleitao]

Uf, amazing. Thanks a lot, @Dandandan . cc @ritchie46 , since this is relevant to Polars.

[jorgecarleitao]

I made two small suggestions just so that we remember why they were added.