Add Stopping Criteria for loop

[Dref360]

Summary:

Allow the users to stop the loop according to various parameters.

Features:

Fixes #243

Checklist:

• Your code is documented (To validate this, add your module to tests/documentation_test.py).
• Your code is tested with unit tests.
• You moved your Issue to the PR state.

[parmidaatg]

Wondering if you wanna change these two lines to below lines to account for general exhaustion of the dataset. The scenario does not apply to this example but as a demonstration of how to account for general exhaustion of the dataset and if the stop criteria is a metric limit

Suggested change

	al_loop.step()
	if stopping_criterion.should_stop():
	flag = al_loop.step()
	if stopping_criterion.should_stop() or flag:

[Dref360]

My thinking was that stopping criteria must be added to ALLoop which would check for both the criteria and exhaustion. It would require ALLoop to know about metrics which it can't do right now.

That would be a major breaking change, so we might want to have a new object instead and deprecate ALLoop 🤔 and then this new object would do the entire experiment?

[parmidaatg]

neat, Just one suggestion but not necessary.

[Dref360]

I'll merge this and then I'll make a proposal for the new object that replaces ActiveLearningLoop into an object that can uses this new Criterion to automatically stop an experiment.

[parmidaatg]

sounds good, lets go