perf: Offer 'simd' feature for faster folding

[epage]

Inspired by https://purplesyringa.moe/blog/i-sped-up-serde-json-strings-by-20-percent/

$ cargo bench && cargo bench -F simd
   Compiling annotate-snippets v0.11.2 (/home/epage/src/personal/annotate-snippets-rs)
    Finished `bench` profile [optimized] target(s) in 0.99s
     Running unittests src/lib.rs (target/release/deps/annotate_snippets-b51bb37991a7f496)

running 0 tests

test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s

     Running benches/bench.rs (target/release/deps/bench-468ba612503afee1)
Timer precision: 18 ns
bench         fastest       │ slowest       │ median        │ mean          │ samples │ iters
├─ fold                     │               │               │               │         │
│  ├─ 0       1.911 µs      │ 19.44 µs      │ 1.943 µs      │ 2.146 µs      │ 100     │ 100
│  ├─ 1       1.916 µs      │ 3.158 µs      │ 1.973 µs      │ 1.982 µs      │ 100     │ 100
│  ├─ 10      2.121 µs      │ 6.05 µs       │ 2.225 µs      │ 2.281 µs      │ 100     │ 100
│  ├─ 100     3.706 µs      │ 7.007 µs      │ 3.83 µs       │ 3.876 µs      │ 100     │ 100
│  ├─ 1000    19.42 µs      │ 25.61 µs      │ 19.48 µs      │ 19.64 µs      │ 100     │ 100
│  ├─ 10000   111.2 µs      │ 204.2 µs      │ 127 µs        │ 133.6 µs      │ 100     │ 100
│  ╰─ 100000  1.094 ms      │ 1.747 ms      │ 1.137 ms      │ 1.158 ms      │ 100     │ 100
╰─ simple     10.14 µs      │ 40.27 µs      │ 10.5 µs       │ 11.01 µs      │ 100     │ 100

   Compiling annotate-snippets v0.11.2 (/home/epage/src/personal/annotate-snippets-rs)
    Finished `bench` profile [optimized] target(s) in 0.99s
     Running unittests src/lib.rs (target/release/deps/annotate_snippets-9d4024ac94675e6a)

running 0 tests

test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s

     Running benches/bench.rs (target/release/deps/bench-d5470149969acbb8)
Timer precision: 13 ns
bench         fastest       │ slowest       │ median        │ mean          │ samples │ iters
├─ fold                     │               │               │               │         │
│  ├─ 0       1.164 µs      │ 13.91 µs      │ 1.208 µs      │ 1.408 µs      │ 100     │ 100
│  ├─ 1       1.188 µs      │ 4.289 µs      │ 1.234 µs      │ 1.277 µs      │ 100     │ 100
│  ├─ 10      1.259 µs      │ 3.822 µs      │ 1.319 µs      │ 1.419 µs      │ 100     │ 100
│  ├─ 100     1.312 µs      │ 2.732 µs      │ 1.412 µs      │ 1.519 µs      │ 100     │ 100
│  ├─ 1000    1.917 µs      │ 5.52 µs       │ 2 µs          │ 2.085 µs      │ 100     │ 100
│  ├─ 10000   7.195 µs      │ 29.55 µs      │ 7.325 µs      │ 7.638 µs      │ 100     │ 100
│  ╰─ 100000  59.08 µs      │ 403 µs        │ 61.1 µs       │ 65.52 µs      │ 100     │ 100
╰─ simple     9.92 µs       │ 19.09 µs      │ 10.33 µs      │ 10.91 µs      │ 100     │ 100

The upper bound for this benchmark was taken from https://github.com/crate-ci/typos/blob/786c825f1753f87b02d24770e3f4ec8043d9084a/crates/typos-dict/src/word_codegen.rs

[Muscraft]

Thanks!