feat(api): include v4 in recent API usage

[elisa-a-v]

Closes #4787

[mlissner]

I didn't check the tests, but the approach looks good to me. Eduardo, I'm going to assign to you because you're helping Elisa get set up, but I don't have a handle on your workload or that of @albertisfu, so if you feel busy, can you guys please discuss and see who makes more sense this sprint? And I'm happy to break a tie if you both feel busy?

[ERosendo]

I think we can use the itertools.batched function in the outer loop to process the data in chunks. This function iterates over the list, yielding tuples of items with a maximum size of n.

Here's the documentation for reference: itertools.batched

Suggested change

	for d, v3_data, v4_data in zip(dates, results[::2], results[1::2]):
	for user_id, count in chain(v3_data, v4_data):
	update_user_counts(user_id, count, d)
	for d, api_usage in zip(dates, batched(results, 2)):
	for user_id, count in chain(*api_usage):
	update_user_counts(user_id, count, d)

This approach enhances future maintainability. If a new API version (e.g., v5) is introduced, we simply need to adjust the batch size instead of modifying the slicing expressions for previous versions (v3, v4) and adding a new slicing expression for v5.

[elisa-a-v]

I added a versions array and used len(versions) instead of a hardcoded 2 so now we'd simply have to add "v5" to the versions array. What do you think? it does add a new nested for loop but it's a fixed and small number of inner iterations (just a couple of versions) so I'm guessing it's not too bad and it's probably worth the tradeoff.

[ERosendo]

@elisa-a-v looks great!