Removed extra copy for transpose arrays in dot()

[antonwolfy]

The PR removes unnecessary copying of input arrays into C-contagious format to call dpnp.dot() function.
The change is indented to speed up execution of code like below:

import dpnp

a = dpnp.ones((128, 4), dtype=dpnp.float32)[::2]
b = dpnp.ones((1048576, 128), dtype=dpnp.float32)
_ = dpnp.dot(a.T, b.T)

This is because gemm() function from MKL blas library supports arrays contiguous on inner dimension, i.e. where a stride for at least one dimension should be equal to 1.

• Have you provided a meaningful PR description?
• Have you added a test, reproducer or referred to 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?
• Have you checked performance impact of proposed changes?
• If this PR is a work in progress, are you filing the PR as a draft?

[vlad-perevezentsev]

LGTM!
thanks @antonwolfy