custom launch and assign kernel names

[bd4]

When profiling and debugging, tracking down which version of a gtensor kernel is the one of interest can be challenging. For SYCL, kernel names are passed as template parameters with templated type names - another sub-type of the templated name could be optionally supplied by the user, via an gt::assign<typename>(lhs, rhs) helper function. This is ugly, but a starting point for discussion. Not exactly sure how to do this for HIP / CUDA yet.