This patch enables vectorization for non-power-of-2 VFs. Initially only
VFs where adding 1 makes the VF a power-of-of-2, i.e. we can still make
relatively effective use of the vectors.

It relies on the existing target cost-models to return accurate costs for
non-power-of-2 vectors. I checked mostly AArch64 and X86 and
there the costs seem reasonable for the costs I checked, although
I expect there will be a need to refine both the cost-models and lowering
to make most effective use of non-power-of-2 SLP vectorization.

Note that re-ordering and shuffling is not implemented for nodes
requiring padding yet to keep the initial implementation simpler.

The feature is guarded by a new flag, off by defaul for now.

also use {} instead of (empty) ResueShuffleIndices.

This is in preparation for
llvm#77790 and makes it easy to add
other, non-power-of-2 VFs for processing.

…-vfs-in-vector

Exit early if MaxVF < MinVF. In that case, the loop body below will
never get entered. Note that this adjusts the condition from MaxVF <=
MinVF. If MaxVF == MinVF, vectorization may still be feasible (and the
loop below gets entered).

… users/fhahn/slp-store-vfs-in-vector

…tor' into slp-vec3

	llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

…-vfs-in-vector

…tor' into slp-vec3