Return default floating point type of the device in dpctl.tensor.astype when newdtype=None

[vlad-perevezentsev]

This PR makes minor changes to dpctl.tensor.astype function by replacing the getting target_dtype.
The old implementation did not consider whether the device where input array is allocated supported double-precision floating point types and returned float64 for None type which raised ValueError.

import dpctl.tensor as dpt

a = dpt.asarray([1, 1, 1], dtype='i4')
dpt.astype(a, None)

ValueError: Device Intel(R) Graphics [0x46a6] does not provide native support 
for double-precision floating point type.

This PR uses _get_dtype function from dpctl.tensor._ctors which fixes wrong behavior.
For None returns default floating point type supported by device.
For all other types returns dpt.dtype(dtype)

• 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?
• Have you checked performance impact of proposed changes?
• If this PR is a work in progress, are you opening the PR as a draft?

[coveralls]

coverage: 83.256% (+0.002%) from 83.254% when pulling 3d51c48 on vlad-perevezentsev:update_astype into 36a7cd7 on IntelPython:master.

[github-actions]

Array API standard conformance tests for dpctl=0.14.5dev0=py310h7bf5fec_1 ran successfully.
Passed: 386
Failed: 614
Skipped: 119