Add missing dtype aliases

[antonwolfy]

The list of missing aliases on numpy types are ported to dpnp:

• cdouble
• complex_
• complexfloating
• cfloat
• csingle
• double
• float_
• floating
• inexact
• int_
• intc
• number
• object_
• signedinteger
• single
• singlecomplex

The flowing type-related functions are added to dpnp:

• issubdtype
• issubsctype

The following constants are borrowed from numpy:

• e
• euler_gamma
• Inf
• inf
• Infinity
• infty
• NAN
• NaN
• NINF
• NZERO
• pi
• PINF
• PZERO

This PR is intended to resolve issue #1307.

• Have you provided a meaningful PR description?
• Have you added a test, reproducer or referred to 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?
• If this PR is a work in progress, are you filing the PR as a draft?

[vlad-perevezentsev]

LGTM!

[oleksandr-pavlyk]

I would suggest to remove dpnp.object_ since dpnp does not support this data type.

[oleksandr-pavlyk]

I do not understand why we export and dpnp.longcomplex (a.k.a np.complex256) as we do not export dpnp.longdouble (a.k.a. np.float128) and we do not support this data type.

We also do no support np.void, so I suggest we remove dpnp.void too.

[antonwolfy]

I do not understand why we export and dpnp.longcomplex (a.k.a np.complex256) as we do not export dpnp.longdouble (a.k.a. np.float128) and we do not support this data type.

We also do no support np.void, so I suggest we remove dpnp.void too.

Ok, agree.
What do you think to add constants like dpnp.pi, dpnp.inf and other ones from numpy? (they are also available in cupy as I can see)

[oleksandr-pavlyk]

pi, e, nan, inf, newaxis are mandated by array API.

[antonwolfy]

All numpy constants ara added to dpnp.