[llvm] [AArch64] Disable consecutive store merging when Neon is unavailable (PR #111519)

[Benjamin Maxwell via llvm-commits]

MacDue wrote:

Yeah, so with `consecutive_stores_quadruple()` for each `@llvm.aarch64.sve.faddv.nxv4f32` you get something like:
```
t33: nxv4f32 = AArch64ISD::FADDV_PRED t20, t10
t35: f32 = extract_vector_elt t33, Constant:i64<0>
```

Which matches `tryStoreMergeOfExtracts()`.

With `consecutive_stores_quadruple2()` for each `@llvm.vector.reduce.fadd.f32.nxv4f32` you get:
```
t20: f32 = vecreduce_fadd nnan ninf nsz arcp contract afn reassoc t8
```
Which does not match any of the store merging patterns.



https://github.com/llvm/llvm-project/pull/111519