Fix ValueError in dpnp.bincount for empty input arrays

[antonwolfy]

dpnp.bincount raised a ValueError when given an empty input array, while NumPy returns an empty (or zero-filled) array. This PR short-circuits the empty-input case so dpnp.bincount matches NumPy.

The Python layer determines the binning edges by reducing over the input with dpnp.max/dpnp.min. Those reductions have no identity for zero-size input, so an empty array failed before ever reaching the kernel:

>>> import dpnp
>>> dpnp.bincount(dpnp.array([], dtype="i8"))
ValueError: reduction cannot be performed over zero-size axes

The C++ kernel already guards sample.get_size() == 0, but that code path is unreachable because the max/min reduction runs first.

The fix assumes to return early for empty input with an intp array of zeros of length minlength, before any reduction is performed.

• 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?
• Have you added documentation for your changes, if necessary?
• Have you added your changes to the changelog?

[github-actions]

View rendered docs @ https://intelpython.github.io/dpnp/pull/3018/index.html

[github-actions]

Array API standard conformance tests for dpnp=0.21.0dev3=py314h509198e_40 ran successfully.
Passed: 1376
Failed: 1
Skipped: 5

[coveralls]

coverage: 78.402% (-0.01%) from 78.414% — fix/bincount-empty-input into master