If we're expanding offsets for a chain of GEPs in RewriteGEPs mode, we
should also rewrite GEPs that have one-use themselves, but are kept
alive by a multi-use GEP later in the chain.

For the sake of simplicity, I've changed this to just skip the one-use
condition entirely (which will perform an unnecessary rewrite of a no
longer used GEP, but shouldn't otherwise matter).

Combine sequences such as:
```llvm
  %pn1 = phi [init1, %BB1], [%op1, %BB2]
  %pn2 = phi [init2, %BB1], [%op2, %BB2]
  %op1 = binop %pn1, constant1
  %op2 = binop %pn2, constant2
  %rdx = binop %op1, %op2
```
Into:
```llvm
  %phi_combined = phi [init_combined, %BB1], [%op_combined, %BB2]
  %rdx_combined = binop %phi_combined, constant_combined
```

This allows us to simplify interleaved reductions, for example as
introduced by the loop vectorizer.

The anecdotal example for this is the loop below:
```c
float foo() {
  float q = 1.f;
  for (int i = 0; i < 1000; ++i)
    q *= .99f;
  return q;
}
```
Which currently gets lowered explicitly such as (on AArch64,
interleaved by four):
```gas
.LBB0_1:
  fmul    v0.4s, v0.4s, v1.4s
  fmul    v2.4s, v2.4s, v1.4s
  fmul    v3.4s, v3.4s, v1.4s
  fmul    v4.4s, v4.4s, v1.4s
  subs    w8, w8, #32
  b.ne    .LBB0_1
```
But with this patch lowers trivially:
```gas
foo:
  mov     w8, #5028
  movk    w8, #14389, lsl #16
  fmov    s0, w8
  ret
```

Co-authored-by: Matt Arsenault <arsenm2@gmail.com>

If we have a gep with vector indices which were splats (either constants
or shuffles), prefer the scalar form of the index. If all operands are
scalarizable, then prefer a scalar gep with splat following.

This does loose some information about undef/poison lanes, but I'm not
sure that's significant versus the number of downstream transformations
which get confused by having to manual scalarize operands.

Detected by CI after #143958.

Extend `isAllocSiteRemovable` to be able to check if the ModRef info
indicates the alloca is only Ref or only Mod, and be able to remove it
accordingly. It seemed that there were a surprising number of
benchmarks with this pattern which weren't getting optimized previously
(due to MemorySSA walk limits). There were somewhat more existing tests
than I'd like to have modified which were simply doing exactly this
pattern (and thus relying on undef memory). Claude code contributed the
new tests (and found an important typo that I'd made).

This implements the discussion in
https://github.com/llvm/llvm-project/pull/143782#discussion_r2142720376.

This simply copies the structure of the vector.reverse patterns from
just above, and reimplements them for the vp.reverse intrinsics when the
mask is all ones and the EVLs exactly match.

Its unfortunate that we have three different ways to represent a reverse
(shuffle, vector.reverse, and vp.reverse) but I don't see an obvious way
to remove any them because the semantics are slightly different.

This significantly improves vectorization in TSVC_2's s112 and s1112
loops when using EVL tail folding.

Seeing how we can't generate any debug intrinsics any more: delete a
variety of codepaths where they're handled. For the most part these are
plain deletions, in others I've tweaked comments to remain coherent, or
added a type to (what was) type-generic-lambdas.

This isn't all the DbgInfoIntrinsic call sites but it's most of the
simple scenarios.

Co-authored-by: Nikita Popov <github@npopov.com>

Fixes https://github.com/llvm/llvm-project/issues/139050.

This patch adds a check to avoid folding min/max reduction into select, which may block loop vectorization.

The issue is that the following snippet:
```
declare i8 @llvm.umin.i8(i8, i8)

define i8 @masked_min_fold_bug(i8 %acc, i8 %val, i8 %mask) {
; CHECK-LABEL: @masked_min_fold_bug(
; CHECK:       %cond = icmp eq i8 %mask, 0
; CHECK:       %masked_val = select i1 %cond, i8 %val, i8 255
; CHECK:       call i8 @llvm.umin.i8(i8 %acc, i8 %masked_val)
;
  %cond = icmp eq i8 %mask, 0
  %masked_val = select i1 %cond, i8 %val, i8 255
  %res = call i8 @llvm.umin.i8(i8 %acc, i8 %masked_val)
  ret i8 %res
}
```

is being optimized to the following code, which can not be vectorized
later.
```
declare i8 @llvm.umin.i8(i8, i8) #0

define i8 @masked_min_fold_bug(i8 %acc, i8 %val, i8 %mask) {
  %cond = icmp eq i8 %mask, 0
  %1 = call i8 @llvm.umin.i8(i8 %acc, i8 %val)
  %res = select i1 %cond, i8 %1, i8 %acc
  ret i8 %res
}

attributes #0 = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
```

Expected:
```
declare i8 @llvm.umin.i8(i8, i8) #0

define i8 @masked_min_fold_bug(i8 %acc, i8 %val, i8 %mask) {
  %cond = icmp eq i8 %mask, 0
  %masked_val = select i1 %cond, i8 %val, i8 -1
  %res = call i8 @llvm.umin.i8(i8 %acc, i8 %masked_val)
  ret i8 %res
}

attributes #0 = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
```

https://godbolt.org/z/cYMheKE5r

Extract a reusable helper for emitting a sum of multiple GEP
offsets.