Rework implementation of `dpnp.fmax` and `dpnp.fmin` functions

The PR proposes to rework implementation of dpnp.fmax and dpnp.fmin functions which are quite limited now.

There is new VM functions are added to use fmax and fmin from OneMKL library where it's possible. Also the PR adds new binary universal functions _fmax and _fmin to _ufunc_impl pybind11 extension which leverages on dpctl f/w.

The all affected tests are updated to cover different use cases.

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

IntelPython / dpnp

Rework implementation of `dpnp.fmax` and `dpnp.fmin` functions #1905