ML_DSA: ntt.rs now has a vectorized structure.

[xvzcf]

No description provided.

[franziskuskiefer]

lgtm, I assume you move other parts in arithmetic later? Or do you first want to vectorise what's in there now and then do more later if necessary, that's fine too.

[xvzcf]

lgtm, I assume you move other parts in arithmetic later? Or do you first want to vectorise what's in there now and then do more later if necessary, that's fine too.

I already starting moving in the arithmetic, the plan is to move every module to a vectorized structure first and then start filling out the AVX2 code and what not after.