Fix AssertionError for numpy.int64 in Variable Initialization

[Hykaruu]

This pull request addresses an issue where an AssertionError is thrown when initializing a Variable with a numpy.int64 type value. The error occurs because the current assertion does not recognize numpy.int64 as a valid integer type.

Proposed Fix
This fix updates the assertion to include numpy.int64 as a valid type. This is done by checking for subtypes of int using numpy.issubdtype.
Additionally, the following code was adapted when casting integers to strings.

Here is the updated code:

assert isinstance(value, (str, bytes, int)) or np.issubdtype(type(value), np.integer), "Value must be a string, int, or image (bytes). Got: {}".format(type(value))
if isinstance(value, int) or np.issubdtype(type(value), np.integer):
    value = str(value)

Testing

• Tested with numpy.int64 values to ensure the assertion no longer fails.

[TranPhu1999]

Tested and worked properly