This stems from a review in PR #53

Ultimately, we want to be able to use the language support for HIP/CUDA rather than bringing in HIP/CUDA through find_package, which is deprecated (though still usable). From the CMake documentation:

It is no longer necessary to use this module or call find_package(CUDA) for compiling CUDA code. Instead, list CUDA among the languages named in the top-level call to the project() command, or call the enable_language() command with CUDA. Then one can add CUDA (.cu) sources directly to targets similar to other languages.

At the moment we are doing a hack overrided the CXX compiler with either hipcc or nvcc.

The reason we're doing it this way (hacky) at the moment is that we get segmentation faults on our AMD GPU tests when using the HIP language support via Cmake, for some yet unknown reason.