[Python][GPU] numba-cuda dependency is maintenance mode only

### Describe the enhancement requested

The [numba-cuda](https://github.com/NVIDIA/numba-cuda) project added a maintenance notice:

> Numba-CUDA is in maintenance mode. Moving forward, we intend to support only security issues and critical bug fixes through the lifetime of CUDA 13. New feature development is targeted towards [Numba-CUDA-MLIR](https://github.com/NVIDIA/numba-cuda-mlir), and we recommend upgrading to Numba-CUDA-MLIR as soon as practical. For migration guidance, see [Migration from Numba / Numba-CUDA](https://github.com/NVIDIA/numba-cuda-mlir#migration-from-numba--numba-cuda).

We should probably deprecate our usage of numba-cuda and migrate to numba-cuda-mlir. Not an expert on this area but using a deprecated maintenance only dependency doesn't sound great.

### Component(s)

GPU, Python

Start by locating the current numba-cuda dependency and the GPU-related Python code that uses it. Read the linked Numba-CUDA-MLIR migration guidance and determine the scope of replacing the dependency; done should mean the supported GPU Python paths and their tests work with the replacement.

Written by the indexing model from the issue text.