Fixed bug in dpctl.tensor.arange, streamlined tests

[oleksandr-pavlyk]

While modifying tests to use dpctl.tensor.arange I encountered exception when starting value was outside of the range of the elemental data-type, e.g. dpctl.tensor.arange(128, 0, -1, dtype='i1').

This PR fixes the issue by coercing the starting point to the elemental data-type scalar.
A test of arange was modified to cover such input.

Tests in test_sycl_event and test_sycl_kernel_submit were modified to use dpctl.tensor.arange as opposed to using numpy.arange and copying the content into a MemoryUSMShared allocation.

• 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?

[github-actions]

View rendered docs @ https://intelpython.github.io/dpctl/pulls/842/index.html

[coveralls]

Coverage increased (+0.002%) to 82.876% when pulling d7fdf41 on use_dpctl_tensor_in_queue_submit_tests into 2522187 on master.

[github-actions]

Deleted rendered PR docs from intelpython.github.com/dpctl, latest should be updated shortly. 🤞