[llvm] AMDGPU: Mark v_mov_b64_pseudo as a VOP1 instruction (PR #128677)

[Matt Arsenault via llvm-commits]

arsenm wrote:

> Seems reasonable. Which part of the remat implementation cares about the VOP1 flag?

canRemat, which I think is just hacking around the default skipping rematerialize due to the exec physreg use 

https://github.com/llvm/llvm-project/pull/128677