[clang] [llvm] [mlir] [AMDGPU] Disable neg_lo[0:1] and neg_hi[0:1] on wmma_f32_16x16x32_bf16 (PR #188649)

[Stanislav Mekhanoshin via cfe-commits]

rampitec wrote:

* #188649 👈
* `main`

<h2></h2>

This stack of pull requests is managed by <a href="https://github.com/shiltian/sgh"><b>sgh</b></a>.
<!-- SGH-MANAGED-COMMENT: DO NOT EDIT OR DELETE THIS COMMENT -->

https://github.com/llvm/llvm-project/pull/188649