[llvm] [bazel] Add a bazel flag to enable building MLIR with CUDA support (PR #88856)

[Adam Paszke via llvm-commits]

apaszke wrote:

> Does this flag correspond to one of the CMake flags?
It corresponds to [this part](https://github.com/llvm/llvm-project/blob/6b6c7e46cc1e97b678e969bad78825dd02c11ff9/mlir/CMakeLists.txt#L115-L119) of the CMake config.

Yes, I have tested the flag locally. It doesn't actually add a dependency on the CUDA toolkit, but only enables a number of MLIR passes that are otherwise `#ifdef`ed away by `MLIR_ENABLE_CUDA_CONVERSIONS`.

https://github.com/llvm/llvm-project/pull/88856