Feature: implement filter for primitive vector (and mask) #5301

Performance Gate Passed

⚡ 15 improvements
✅ 1347 untouched
🆕 43 new
⏩ 567 skipped¹

	Benchmark	`BASE`	`HEAD`	Change
⚡	`take_indices[(1000, 16, 0.005)]`	22.3 µs	19.5 µs	+14.56%
⚡	`take_indices[(1000, 16, 0.01)]`	22.4 µs	19.6 µs	+14.29%
⚡	`take_indices[(1000, 16, 0.03)]`	24.4 µs	21.4 µs	+14.18%
⚡	`take_indices[(1000, 256, 0.005)]`	22.1 µs	19.2 µs	+14.75%
⚡	`take_indices[(1000, 256, 0.01)]`	22.3 µs	19.5 µs	+14.43%
⚡	`take_indices[(1000, 256, 0.03)]`	22.7 µs	19.9 µs	+14.13%
⚡	`take_indices[(1000, 4, 0.005)]`	24.9 µs	21.8 µs	+14.01%
⚡	`take_indices[(1000, 4, 0.01)]`	23.7 µs	20.6 µs	+14.67%
⚡	`take_indices[(1000, 4, 0.03)]`	26.7 µs	23.7 µs	+12.8%
⚡	`take_indices[(10000, 16, 0.005)]`	26 µs	23 µs	+13.19%
⚡	`take_indices[(10000, 16, 0.01)]`	28.9 µs	25.8 µs	+11.73%
⚡	`take_indices[(10000, 256, 0.005)]`	24.3 µs	21.5 µs	+13.08%
⚡	`take_indices[(10000, 256, 0.01)]`	26.3 µs	23.5 µs	+11.95%
⚡	`take_indices[(10000, 4, 0.005)]`	28.6 µs	25.6 µs	+11.84%
⚡	`take_indices[(10000, 4, 0.01)]`	32.4 µs	29.4 µs	+10.32%
🆕	`expand_buffer[u32, (1024, 0.1)]`	N/A	8.2 µs	N/A
🆕	`expand_buffer[u32, (1024, 0.5)]`	N/A	9.5 µs	N/A
🆕	`expand_buffer[u32, (1024, 0.9)]`	N/A	10.4 µs	N/A
🆕	`expand_buffer[u32, (16384, 0.1)]`	N/A	62.4 µs	N/A
🆕	`expand_buffer[u32, (16384, 0.5)]`	N/A	83.2 µs	N/A
...	...	...	...	...

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

567 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩