This PR changes a bit the behavior for dpctl.tensor.sum function for scalars.
The previous implementation always returned dpt.zeros for zero-dimensions of the input array and worked incorrectly for scalars. According to the Python Array API: If N is 0, the sum is 0 (i.e., the empty sum) [sum].
Now if dpctl.tensor.sum takes a scalar it returns a scalar has the same behavior as numpy

#old impl
a = dpt.asarray(1)
dpt.sum(a)
usm_ndarray(0., dtype=float32)

a = dpt.asarray(())
dpt.sum(a)
usm_ndarray(0., dtype=float32)

#new impl
a = dpt.asarray(1)
dpt.sum(a)
usm_ndarray(1., dtype=float32)

a = dpt.asarray(())
dpt.sum(a)
usm_ndarray(0., dtype=float32)

• 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?