Add explicit SIMD for aarch64 platforms.

[cessen]

Aarch64 simd intrinsics are stable as of Rust 1.59 (although due to a documentation bug are falsely still marked as unstable: rust-lang/stdarch#1268). This means we can now add proper SIMD acceleration for those platforms.

I don't currently have an aarch64 system to test on, so I won't attempt to implement myself. But if anyone would like to take a crack at it, it mostly just consists of making a new impl of the ByteChunk trait in src/byte_chunk.rs for the appropriate core::arch::aarch64 type. Although there may be some complexities with CPU features, etc.

[CeleritasCelery]

I took the initiative to implement this on Aarch64 on my M1.

CeleritasCelery@2757d69

The tests pass and the fuzzer didn't find anything, but this code is significantly slower then the previous non-SIMD version 😢.

here is the results. We have 29 improvements (around 2x), and 79 regressions (2-10x).

I am at a loss to explain the slow down. Would you be able to take a look and see if I got something fundamentally wrong?

[cessen]

Thanks so much for taking a crack at this!

If you just want to submit a WIP PR directly, that would be fine. Even if it isn't up to snuff yet, that will make reviewing it a little easier.

Either way, I'll try to get to it within the next couple of weeks.