The target is called "x86" not "x86_64".

Fixes #145511

Note that there are still two instances of
--passes=slp-vectorizer,instcombine left unchanged because it seems that
the tests are meant to run in conjunction with instcombine and removing
instcombine would invalidate their original objective:


[llvm/test/Transforms/SLPVectorizer/arith-div-undef.ll](https://github.com/llvm/llvm-project/blob/main/llvm/test/Transforms/SLPVectorizer/arith-div-undef.ll)

[llvm/test/Transforms/SLPVectorizer/slp-hr-with-reuse.ll](https://github.com/llvm/llvm-project/blob/main/llvm/test/Transforms/SLPVectorizer/slp-hr-with-reuse.ll)

This canonicalizes fneg/fabs (shuffle X, poison, mask) -> shuffle
(fneg/fabs X), posion, mask

This undoes part of b331a7ebc1e02f9939d1a4a1509e7eb6cdda3d38 and
a8f13dbdeb31be37ee15b5febb7cc2137bbece67, but keeps the binary shuffle
case i.e. shuffle fneg, fneg, mask.

By pulling out the shuffle we bring it inline with the same
canonicalisation we perform on binary ops and intrinsics, which the
original commit acknowledges it goes in the opposite direction.

However nowadays VectorCombine is more powerful and can do more
optimisations when the shuffle is pulled out, so I think we should
revisit this. In particular we get more shuffles folded and can perform
scalarization.

This patch adjusts the cost model to account for the ability of the
AMDGPU optimizer to group together i8 values into i32 values.

Co-authored-by: Erich Keane <ekeane@nvidia.com>

X86 uses implicit widening and BLEND/MOV shuffles in these cases - otherwise we still treat it as a SK_PermuteTwoSrc

The memory value and mask value types might legalise differently - e.g. a v64i32 might split into 4 x v16i32 / 8 x v8i32 but the mask might legalize as 1 x v64i8 / 2 x v32i8 etc.

If the legalised value type has been split, then we must ensure we compute the cost for the entire mask value type and let getShuffleCost handle any legalisation, not assume that only a single trailing split mask will require widening.

If a SK_PermuteTwoSrc shuffle kind's mask only references the first
operand, then treat this as SK_PermuteSingleSrc

Part of #145335

We should only divide the number of pieces to fit the packed instructions
if we actually have pk instructions. This increases the cost of copysign,
but is closer to the current codegen output. It could be much cheaper
than it is now.

Patch fixes cost estimation for the extractelements from non-power-of-2
vectors, defined as subvector extracts. In this case the subvector size
might be not adjusted to a whole register size, need to get the minimum
between whole vector size and the actual difference to prevent compiler
crash.

Fixes #143513