Support tests on single point precision GPUs

[ZzEeKkAa]

Most of the integrated GPUs does not support doble precision and numba by default uses double precision floats. That makes our tests fail on those GPUs. This changes targeted to add support for tests on single precision GPUs by:

• tests that where using default allocation or only float64/complex128 uses were patched to use float32/complex64 implementations;
• tests that accept dtype parameters where patched to run only on supported types
• few issues where found with lowering or silent crashes. Those tests were patched to bypass these cases.

Discovered issues:

• arctan2 does not work for float32 if device does not support float64 (silent crash)
• parfor_divide and parfor_exponent does not work on int32 and int64 if device does not support float64 (lowering error)

Issue checks:

• Have you provided a meaningful PR description?
• Have you added a test, reproducer or referred to an issue with a reproducer?
• Have you tested your changes locally for CPU and GPU devices?
• Have you made sure that new changes do not introduce compiler warnings?
• If this PR is a work in progress, are you filing the PR as a draft?

[diptorupd]

Thank you. Great changes. Please merge after CI is green