Add bits as an optional feature for bitvec

[wcampbell0x2a]

This is like the final boss of all my performance work. I try to explain the "no warranty and testing are given for the disable of this feature", while still implementing enough to make this useful in the case of "I want perf to matter".

bench: https://github.com/wcampbell0x2a/deku-bench/pull/24

[github-actions[bot]]

Benchmark for 4a76c64

Click to view benchmark

| Test | Base | PR | % | |------|--------------|------------------|---| | count | 3.7±0.04µs | N/A | N/A | | deku_read_bits | 1320.4±7.58ns | **1261.2±10.48ns** | **-4.48%** | | deku_read_byte | 3.4±0.22ns | **3.3±0.08ns** | **-2.94%** | | deku_read_enum | **2.6±0.06ns** | 3.0±0.06ns | **+15.38%** | | deku_read_vec | 35.5±2.29ns | 35.2±0.54ns | -0.85% | | deku_write_bits | **149.4±1.91ns** | 168.0±6.70ns | **+12.45%** | | deku_write_byte | 22.0±0.28ns | 21.9±0.29ns | -0.45% | | deku_write_enum | 21.4±0.22ns | 21.4±0.46ns | 0.00% | | deku_write_vec | 387.5±4.44ns | **330.7±5.39ns** | **-14.66%** | | read_all | 4.4±0.18µs | N/A | N/A | | read_all_bytes | 3.9±0.03µs | N/A | N/A |

[github-actions[bot]]

Benchmark for 8947ee1

Click to view benchmark

| Test | Base | PR | % | |------|--------------|------------------|---| | count | 3.7±0.04µs | N/A | N/A | | deku_read_bits | 1395.3±21.80ns | **1285.8±42.23ns** | **-7.85%** | | deku_read_byte | 3.3±0.06ns | 3.3±0.05ns | 0.00% | | deku_read_enum | **2.6±0.06ns** | 3.0±0.06ns | **+15.38%** | | deku_read_vec | 35.4±0.51ns | 35.3±0.45ns | -0.28% | | deku_write_bits | 161.1±5.19ns | **146.4±4.32ns** | **-9.12%** | | deku_write_byte | 22.1±0.41ns | 21.8±0.39ns | -1.36% | | deku_write_enum | 21.4±0.23ns | 21.4±0.41ns | 0.00% | | deku_write_vec | 315.0±6.45ns | 313.0±4.11ns | -0.63% | | read_all | 4.4±0.06µs | N/A | N/A | | read_all_bytes | 3.9±0.06µs | N/A | N/A |

[github-actions[bot]]

Benchmark for 755ed48

Click to view benchmark

| Test | Base | PR | % | |------|--------------|------------------|---| | count | 3.7±0.03µs | N/A | N/A | | deku_read_bits | 1375.8±19.60ns | **1303.6±38.98ns** | **-5.25%** | | deku_read_byte | 3.5±0.06ns | **3.2±0.04ns** | **-8.57%** | | deku_read_enum | 3.0±0.09ns | **2.6±0.05ns** | **-13.33%** | | deku_read_vec | 38.1±2.11ns | **35.6±0.40ns** | **-6.56%** | | deku_write_bits | **160.4±4.86ns** | 175.5±5.04ns | **+9.41%** | | deku_write_byte | 22.2±0.28ns | **21.8±0.39ns** | **-1.80%** | | deku_write_enum | 21.4±0.19ns | 21.4±0.45ns | 0.00% | | deku_write_vec | **313.3±10.86ns** | 412.2±4.90ns | **+31.57%** | | read_all | 4.2±0.06µs | N/A | N/A | | read_all_bytes | 5.0±0.04µs | N/A | N/A |

[github-actions[bot]]

Benchmark for 6bea42b

Click to view benchmark

| Test | Base | PR | % | |------|--------------|------------------|---| | count | 3.7±0.04µs | 3.7±0.03µs | 0.00% | | deku_read_bits | 1273.2±21.85ns | 1285.2±21.64ns | +0.94% | | deku_read_byte | 3.6±0.07ns | **3.2±0.05ns** | **-11.11%** | | deku_read_enum | 3.0±0.04ns | **2.6±0.06ns** | **-13.33%** | | deku_read_vec | 35.7±0.53ns | 35.4±0.60ns | -0.84% | | deku_write_bits | **150.4±2.19ns** | 160.7±6.36ns | **+6.85%** | | deku_write_byte | 21.9±0.41ns | 22.0±0.22ns | +0.46% | | deku_write_enum | 21.4±0.19ns | 21.4±0.37ns | 0.00% | | deku_write_vec | **329.4±4.28ns** | 374.2±4.39ns | **+13.60%** | | read_all | 4.4±0.04µs | **4.2±0.05µs** | **-4.55%** | | read_all_bytes | **3.9±0.05µs** | 5.0±0.03µs | **+28.21%** |