Convert numpy.int64 to python native int in 'random_scale'

[smarginatura]

This pull request addresses a TypeError triggered by the random_scale function in transform.py

The error is encountered when running certain Colab notebooks based on lucent ( (eg:neuron_interaction) with PyTorch 2.1.

The issue is resolved by converting the scale_shape local variable from a list of numpy.int64 to native Python int, ensuring compatibility with nn.Upsample.