dpctl.tensor.reshape raises ValueError with empty list as input

[LukichevaPolina]

Reproduces:

import dpctl.tensor as dpt

a = dpt.asarray([])
dpt.reshape(a, (0,))

Output:

Traceback (most recent call last):
  File "/localdisk/work/plukiche/dpnp_fork/dpctl_empty_reshape_repro.py", line 4, in <module>
    dpt.reshape(a, (0,))
  File "/localdisk/work/plukiche/miniconda/miniconda3/envs/dpnptc4_internal/lib/python3.8/site-packages/dpctl/tensor/_reshape.py", line 107, in reshape
    newsts = reshaped_strides(X.shape, X.strides, newshape, order=order)
  File "/localdisk/work/plukiche/miniconda/miniconda3/envs/dpnptc4_internal/lib/python3.8/site-packages/dpctl/tensor/_reshape.py", line 50, in reshaped_strides
    for flat_index in [
  File "/localdisk/work/plukiche/miniconda/miniconda3/envs/dpnptc4_internal/lib/python3.8/site-packages/dpctl/tensor/_reshape.py", line 51, in <listcomp>
    np.ravel_multi_index(unitvec, new_sh, order=order)
  File "<__array_function__ internals>", line 5, in ravel_multi_index
ValueError: cannot unravel if shape has zero entries (is empty).

dpctl version is 0.12.0dev1+104.gccd21b68

[LukichevaPolina]

@oleksandr-pavlyk, Could you please look at this problem?

[densmirn]

@oleksandr-pavlyk, could you please either leave a comment or prepare a fix on this issue? The issue is blocking DPNP testing in external CI.

[oleksandr-pavlyk]

The immediate error is caused by attempting to use np.ravel_multi_index for shape with zero elements. That case should be handled separately.