Filter DecimalArray as PrimitiveArray ~5x Faster (#2637)

[tustvold]

Which issue does this PR close?

Part of #2637

Rationale for this change

Uses the optimised filter primitive implementation for filtering decimal arrays

What changes are included in this PR?

Are there any user-facing changes?

No

[Dandandan]

LGTM - would be interested what is the performance change

[tustvold]

filter decimal128 (kept 1/2)
                        time:   [92.406 µs 92.455 µs 92.515 µs]
                        change: [-66.107% -66.043% -65.967%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  4 (4.00%) high mild
  7 (7.00%) high severe

filter decimal128 high selectivity (kept 1023/1024)
                        time:   [51.903 µs 52.226 µs 52.571 µs]
                        change: [-58.848% -58.570% -58.303%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild

filter decimal128 low selectivity (kept 1/1024)
                        time:   [2.8865 µs 2.8877 µs 2.8891 µs]
                        change: [-20.743% -20.519% -20.286%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high severe

filter context decimal128 (kept 1/2)
                        time:   [41.956 µs 43.880 µs 45.982 µs]
                        change: [-84.841% -84.535% -84.193%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
  13 (13.00%) high severe

filter context decimal128 high selectivity (kept 1023/1024)
                        time:   [49.703 µs 50.000 µs 50.358 µs]
                        change: [-58.433% -58.128% -57.825%] (p = 0.00 < 0.05)
                        Performance has improved.

filter context decimal128 low selectivity (kept 1/1024)
                        time:   [367.45 ns 367.51 ns 367.59 ns]
                        change: [-88.839% -88.822% -88.791%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  1 (1.00%) low mild
  6 (6.00%) high severe

[ursabot]

Benchmark runs are scheduled for baseline = f055f51 and contender = ede36d7. ede36d7 is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ec2-t3-xlarge-us-east-2] ec2-t3-xlarge-us-east-2
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on test-mac-arm] test-mac-arm
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-i9-9960x] ursa-i9-9960x
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-thinkcentre-m75q] ursa-thinkcentre-m75q
Buildkite builds:
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java

[Dandandan]

Nice @tustvold