[llvm] [RISCV] Disable fixed vectors in getOptimalMemOpType if minimum VLEN is 32. (PR #102974)

[Huaqi Fang via llvm-commits]

fanghuaqi wrote:

> > At the moment, trying to use zve32x/zve32f is living very dangerously.
> 
> zve32x/zve32f are supposed to work as long as zvl64b or higher is also used.

Yes, I tested with my case, when I pass `_zve32f` it will not generate any vector instruction, and I want to ask how can I disable the auto vectorization for RVV since most of my cases are written by using rvv intrinsic API.

https://github.com/llvm/llvm-project/pull/102974