New PerlinNoiseAV test in SMHasher

Sorry to write this directly as I'm an "interested person" having my own hash functions, but recently the claim "xxhash passes all SMHasher tests" became imprecise: I've invented a new Perlin Noise test that finds "holes" in several hashes, not just xxhash. Maybe you can fix this as xxhash is too popular to miss that fact. I'm not expecting to gain anything from my post; my `komihash` hash function is relatively new, so needs a lot more time to gain traction in this very competitive "hash function" field.

Maybe you can also add `komihash` to your tests, it's probably not as fast as xxhash3, though, but useful as a comparison to the "state-of-the-art" hash functions. (`wyhash` comparison would be great as well)

https://github.com/avaneev/komihash
https://github.com/rurban/smhasher

Start by reviewing the SMHasher test suite and the PerlinNoiseAV results described in the issue, then inspect the linked komihash and smhasher repositories for comparison context. Done means deciding how the new test affects the xxHash claim and adding the requested hash-function comparisons if the project accepts them.

Written by the indexing model from the issue text.