Alexey Bader

This PR should be ready for merge. The non-blocking issues in the front-end can be addressed separately. @tahonermann, @aelovikov-intel, could you please approve if there are no other comments?

`SYCL :: Reduction/reduction_nd_lambda.cpp` failure on Windows is not related to the patch. This issue is tracked by https://github.com/intel/llvm/issues/19354.

> * A change to diagnose declarations that have both the `sycl_device` and `sycl_device_only` attributes as errors. > * A change to diagnose overload sets that contain both `sycl_device` and...

Tom has approved this PR, but apparently, he is not member of the @intel/dpcpp-cfe-reviewers team. @intel/dpcpp-cfe-reviewers, could someone approve to unblock the merge, please?

@npmiller, a couple of follow-ups. - I think we should drop `-fallback` from the `cmath-fallback.cpp` test name. - ~`libdevice` incremental build is broken. When I make change in the `sycl/stl_wrappers/cmath`...

> > Just to clarify this PR implements both the "wrappers" but also the "fallback" in one step, it doesn't need anything else from libdevice. So in theory enabling this...