Benchmark fused comparison for all widths

[AdamGS]

Enable fused comparison benchmarking, and run it for all widths

[AdamGS]

This is a divan limitations, if used in a const generic it can't be more than 20 items.

[codspeed-hq]

Merging this PR will degrade performance by 18.72%

❌ 1 regressed benchmark
✅ 26 untouched benchmarks
🆕 120 new benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	pack_16_to_3_stack	1.8 µs	2.2 µs	-18.72%
🆕	bitpacking_cmp_fused[u32, 16]	N/A	3.6 µs	N/A
🆕	bitpacking_cmp_fused[u32, 19]	N/A	4 µs	N/A
🆕	bitpacking_cmp_fused[u32, 18]	N/A	3.9 µs	N/A
🆕	bitpacking_cmp_fused[u32, 17]	N/A	3.8 µs	N/A
🆕	bitpacking_cmp_fused[u32, 20]	N/A	4 µs	N/A
🆕	bitpacking_cmp_fused[u32, 6]	N/A	3.1 µs	N/A
🆕	bitpacking_cmp_fused[u32, 3]	N/A	2.8 µs	N/A
🆕	bitpacking_cmp_fused[u32, 21]	N/A	4.2 µs	N/A
🆕	bitpacking_cmp_fused[u32, 30]	N/A	4.8 µs	N/A
🆕	bitpacking_cmp_fused[u16, 12]	N/A	2.6 µs	N/A
🆕	bitpacking_cmp_fused[u16, 11]	N/A	2.6 µs	N/A
🆕	bitpacking_cmp_fused[u16, 10]	N/A	2.6 µs	N/A
🆕	bitpacking_cmp_fused[u32, 22]	N/A	4.2 µs	N/A
🆕	bitpacking_cmp_fused[u32, 31]	N/A	4 µs	N/A
🆕	bitpacking_cmp_fused[u32, 24]	N/A	4.2 µs	N/A
🆕	bitpacking_cmp_fused[u16, 13]	N/A	2.7 µs	N/A
🆕	bitpacking_cmp_fused[u32, 4]	N/A	2.8 µs	N/A
🆕	bitpacking_cmp_fused[u16, 2]	N/A	2.3 µs	N/A
🆕	bitpacking_cmp_fused[u32, 5]	N/A	3 µs	N/A
...	...	...	...	...

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

---

_{Comparing adamg/benchmark-fuze-compare-all-widths (237049e) with develop (f6a7c0e)}