Add missing type hints and make them consistent

[abdelrahman725]

1. It would be nice to add the missing type hints to instnace attributes and functions return type. This improves readability and development. It will also avoid the vague Unknown type when hovering over code inside code editors like VS code.

2. Consistnecy: some annotations are not consistent, mixing built-in generics with typing generics (e.g., Dict and dict). Since Experanto supports Python>=3.9 (as defined in pyproject.toml), I believe it will be safe to migrate to built-in types.

I've already started working on this, but I'm opening this issue first to follow what Polina suggested in GSoC forum. I will open a PR once you approve

[pollytur]

@abdelrahman725 could you please link a couple of functions where the type hints were missing and it was confusing. I am a bit surprised pyright did not caught it

[abdelrahman725]

@pollytur

Here are some :

• get_data_key_from_root_folder missing return type.

• self.trials here should be annotated as list[ScreenTrial], not only for the sake of this variable but also for self._num_frames and self._num_frames (here) which consume it. Currently they are shown by VS code as list[Unknown]. I've tested my changes, and the expected types appear.

• SequenceInterpolator has some instance attributes with type Any. This is not optimal, as you can see self.time_delta expectes self.sampling_rate to be a number (float or int) by doing the following: self.time_delta = 1.0 / self.sampling_rate

Also, we should consider the consistency aspect I mentioned above

[pollytur]

ok, lets add the missing type hints where needed. Though we are moving up to python 3.10, so use these type notations
As for changing the old ones for the sake of consistency - I might suggest to skip it for now as I do not see the added value on it.

I have assigned you for this issue for now but please tackle it after the allen example and BWM datasets