Better fixed-width buffer filtering

[robert3005]

Port logic from Primitive types to Decimal types and adjust the way we choose
filter strategies from simple threshold for indices and slices

[codspeed-hq]

Merging this PR will improve performance by 43.16%

⚡ 13 improved benchmarks
❌ 2 regressed benchmarks
✅ 1653 untouched benchmarks
⏩ 38 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	slice_empty_vortex	339.4 ns	397.8 ns	-14.66%
❌	Simulation	chunked_varbinview_opt_canonical_into[(100, 100)]	305.4 µs	340.6 µs	-10.34%
⚡	Simulation	patterns_i128[Contiguous]	234.3 µs	17.9 µs	×13
⚡	Simulation	random_i64[0.5]	204.1 µs	132 µs	+54.64%
⚡	Simulation	random_i64[0.8]	277.3 µs	186 µs	+49.08%
⚡	Simulation	patterns_i128[Runs]	244.5 µs	174 µs	+40.53%
⚡	Simulation	random_i128[0.8]	388.8 µs	277.5 µs	+40.11%
⚡	Simulation	patterns_i128[Random]	290.1 µs	219.4 µs	+32.24%
⚡	Simulation	random_i128[0.5]	289.4 µs	219.5 µs	+31.81%
⚡	Simulation	random_i256[0.5]	474.5 µs	375.2 µs	+26.48%
⚡	Simulation	chunked_varbinview_canonical_into[(1000, 10)]	190.9 µs	154.5 µs	+23.51%
⚡	Simulation	random_i256[0.8]	646.1 µs	527.3 µs	+22.53%
⚡	Simulation	bitwise_not_vortex_buffer_mut[128]	215.3 ns	186.1 ns	+15.67%
⚡	Simulation	bitwise_not_vortex_buffer_mut[1024]	275.6 ns	246.4 ns	+11.84%
⚡	Simulation	compact[(4096, 90)]	896.4 ns	808.9 ns	+10.82%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing rk/filterbuffer (8633f26) with develop (0c7e0ff)}

Footnotes

1. 38 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. ↩

[robert3005]

Moved benchmarks here #8727