`is_int` check fails for Numpy integers

[ThibeauWouters]

The is_int check function defined here will fail in case the argument is a Numpy integer (numpy version 1.26.4):

>>> my_int = np.int64(6)
>>> type(my_int)
<class 'numpy.int64'>
>>> isinstance(my_int, int)
False

This will interrupt a program if it generates a Numpy integer rather than regular Python integer. This can easily be resolved by importing Numpy in the utils/typechecks.py script and changing

return isinstance(x, int)

to, e.g.

return isinstance(x, (int, np.integer))

Alternatively, one can call int(x) before doing the check. I can create a PR to implement this easy fix if the developers tell me which solution would be preferable.