This patch removes some of the benchmarks in our suite that are
relatively low value. Specifically:
- `lexicographical_compare` isn't benchmarked with every number of
elements between 1 and 8 anymore
- `bitset` isn't benchmarked with as many sizes anymore. Instead, only
some representative sizes are benchmarked.
- `to_chars` isn't benchmarked for every possible base anymore, but only
2, 8, 10, 16 and 23 (so we have an uncommon base as well)
It's important for the name of each benchmark to fully describe the
benchmark since we aggregate the results of all benchmarks in a single
.lnt file for analysis.
This PR simplifies `__bitset::__init` into a more compact and readable
form, which avoids redundant computations of a `size_t` value and
eliminates the overhead of a local array.
This patch optimizes `bitset::to_string` by replacing the existing bit-by-bit processing with a more efficient
bit traversal strategy. Instead of checking each bit sequentially, we leverage `std::__countr_zero` to efficiently
locate the next set bit, skipping over consecutive zero bits. This greatly accelerates the conversion process,
especially for sparse `bitset`s where zero bits dominate. To ensure similar improvements for dense `bitset`s, we
exploit symmetry by inverting the bit pattern, allowing us to apply the same optimized traversal technique. Even
for uniformly distributed `bitset`s, the proposed approach offers measurable performance gains over the existing
implementation.
Benchmarks demonstrate substantial improvements, achieving up to 13.5x speedup for sparse `bitset`s with
`Pr(true bit) = 0.1`, 16.1x for dense `bitset`s with `Pr(true bit) = 0.9`, and 8.3x for uniformly distributed
`bitset`s with `Pr(true bit) = 0.5)`.
