[llvm] [GISel] Add support for scalable vectors in getLCMType (PR #80306)

[Michael Maitland via llvm-commits]

michaelmaitland wrote:

> G_UNMERGE_VALUES, G_MERGE_VALUES, and G_BUILD_VECTOR only work with fixed length vectors. Is the root cause of your crash somewhere else?

I am not so sure that I agree with `G_UNMERGE_VALUES, G_MERGE_VALUES only work with fixed length vectors`. Check this test out:
```
; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 4
; RUN: llc -mtriple=riscv32 -mattr=+v -global-isel -stop-after=irtranslator \
; RUN:   -verify-machineinstrs < %s | FileCheck -check-prefixes=RV32 %s
; RUN: llc -mtriple=riscv64 -mattr=+v -global-isel -stop-after=irtranslator \
; RUN:   -verify-machineinstrs < %s | FileCheck -check-prefixes=RV64 %s
define <vscale x 16 x i64> @shufflevector_nxv16i64_1() {
  ; RV32-LABEL: name: shufflevector_nxv16i64_1
  ; RV32: bb.1 (%ir-block.0):
  ; RV32-NEXT:   [[DEF:%[0-9]+]]:_(<vscale x 16 x s64>) = G_IMPLICIT_DEF
  ; RV32-NEXT:   [[SHUF:%[0-9]+]]:_(<vscale x 16 x s64>) = G_SHUFFLE_VECTOR [[DEF]](<vscale x 16 x s64>), [[DEF]], hufflemask(undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef)
  ; RV32-NEXT:   [[UV:%[0-9]+]]:_(<vscale x 8 x s64>), [[UV1:%[0-9]+]]:_(<vscale x 8 x s64>) = G_UNMERGE_VALUES [SHUF]](<vscale x 16 x s64>)
  ; RV32-NEXT:   $v8m8 = COPY [[UV]](<vscale x 8 x s64>)
  ; RV32-NEXT:   $v16m8 = COPY [[UV1]](<vscale x 8 x s64>)
  ; RV32-NEXT:   PseudoRET implicit $v8m8, implicit $v16m8
  ;
  ; RV64-LABEL: name: shufflevector_nxv16i64_1
  ; RV64: bb.1 (%ir-block.0):
  ; RV64-NEXT:   [[DEF:%[0-9]+]]:_(<vscale x 16 x s64>) = G_IMPLICIT_DEF
  ; RV64-NEXT:   [[SHUF:%[0-9]+]]:_(<vscale x 16 x s64>) = G_SHUFFLE_VECTOR [[DEF]](<vscale x 16 x s64>), [[DEF]], shufflemask(undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef, undef)
  ; RV64-NEXT:   [[UV:%[0-9]+]]:_(<vscale x 8 x s64>), [[UV1:%[0-9]+]]:_(<vscale x 8 x s64>) = G_UNMERGE_VALUES [[SHUF]](<vscale x 16 x s64>)
  ; RV64-NEXT:   $v8m8 = COPY [[UV]](<vscale x 8 x s64>)
  ; RV64-NEXT:   $v16m8 = COPY [[UV1]](<vscale x 8 x s64>)
  ; RV64-NEXT:   PseudoRET implicit $v8m8, implicit $v16m8
  %a = shufflevector <vscale x 16 x i64> undef, <vscale x 16 x i64> undef, <vscale x 16 x i32> undef
  ret <vscale x 16 x i64> %a
}
```

There is definitely G_UNMERGE_VALUES getting generated. This test was generated with my fixing of get LCM/GCD types updated. 

https://github.com/llvm/llvm-project/pull/80306