[llvm] [GlobalISel] Fix commute_shift_binop apply to use MachineIRBuilder (PR #222482)

[Vikash Gupta via llvm-commits]

vg0204 wrote:

The MIR-patttern also supports binary constant folding via `constant_fold_binop` CombineRule that supposed to do G_LSHL folding, but the real reason its not happening is that it consumes  vector constants which is  not handled yet, so https://github.com/llvm/llvm-project/pull/220876 adds support for that precisely.

https://github.com/llvm/llvm-project/pull/222482