Optimized implementation for uN::{gather,scatter}_bits

[quaternic]

Feature gate: #![feature(uint_gather_scatter_bits)]
Tracking issue: #149069
Accepted ACP: rust-lang/libs-team#695

Implements the methods using the parallel suffix strategy mentioned in the ACP discussion. The referenced source material provides C implementations, though this PR makes improvements over those, cutting the instruction count by a third:
https://rust.godbolt.org/z/rn5naYnK4 (this PR)
https://c.godbolt.org/z/WzYd5WbsY (Hacker's delight)

This was initially based on the code for gather_bits that @okaneco provided in rust-lang/libs-team#695 (comment) . I wanted to understand how it worked, and later on noticed some opportunities for improvement, which eventually led to this PR.

[quaternic]

Do take this with a grain of salt, since I authored the benchmarks with this in mind. In particular, since the new implementation doesn't have any input-dependent control-flow, it is easily vectorized which all of these benchmarks allow for.

Benchmarked locally on an Intel Core i7 920 @ 2.67GHz (from ~2009)

	old (ns/iter)	new (ns/iter)	old/new
num::int_bits::u8::constant::gather_bits	134	193	0.69
num::int_bits::u8::constant::scatter_bits	133	143	0.93
num::int_bits::u8::invariant::gather_bits	8723	193	45.11
num::int_bits::u8::invariant::scatter_bits	10827	187	57.85
num::int_bits::u8::variable::gather_bits	17938	734	24.43
num::int_bits::u8::variable::scatter_bits	19809	838	23.64
num::int_bits::u16::constant::gather_bits	278	290	0.96
num::int_bits::u16::constant::scatter_bits	279	198	1.41
num::int_bits::u16::invariant::gather_bits	9401	231	40.66
num::int_bits::u16::invariant::scatter_bits	9480	233	40.71
num::int_bits::u16::variable::gather_bits	16398	932	17.59
num::int_bits::u16::variable::scatter_bits	14793	1083	13.66
num::int_bits::u32::constant::gather_bits	528	373	1.42
num::int_bits::u32::constant::scatter_bits	520	301	1.73
num::int_bits::u32::invariant::gather_bits	7699	284	27.11
num::int_bits::u32::invariant::scatter_bits	6670	295	22.6
num::int_bits::u32::variable::gather_bits	9993	1394	7.17
num::int_bits::u32::variable::scatter_bits	9051	1620	5.59
num::int_bits::u64::constant::gather_bits	1008	387	2.6
num::int_bits::u64::constant::scatter_bits	1015	377	2.69
num::int_bits::u64::invariant::gather_bits	7892	347	22.78
num::int_bits::u64::invariant::scatter_bits	6730	362	18.58
num::int_bits::u64::variable::gather_bits	8930	1930	4.63
num::int_bits::u64::variable::scatter_bits	7981	2238	3.57
num::int_bits::u128::constant::gather_bits	15690	721	21.77
num::int_bits::u128::constant::scatter_bits	11374	655	17.35
num::int_bits::u128::invariant::gather_bits	16542	856	19.33
num::int_bits::u128::invariant::scatter_bits	13316	864	15.4
num::int_bits::u128::variable::gather_bits	16984	4403	3.86
num::int_bits::u128::variable::scatter_bits	13721	4674	2.94

[rustbot]

r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer