fetch-data: Planning for pagination

Pagination is a complex issue to deal with. There are several scenarios:

* Retrieve a page at a time, and track page offset + max pages
* Infinite scroll, where the first page is loaded, and subsequent ones are loaded while scrolling
* Large infinite scoll, where previous data has to be cleared as the list scrolls
* Load everything, for times when you just want to take the hit and load it all, but should there be a limit? And what if it's reached?

In addition, there are other concerns:

* What if data on a previous page changes while you're scrolling? Will it change the offset?
* What about dependent data loaded for each element?
* What about partial data loads?

This ticket is just a placeholder for the above strategies and obstacles. More tickets will need to be added to implement each one.

No files, tests, or entry points are named. Start by reviewing the listed pagination strategies and obstacles, then separate them into focused implementation tickets. Done for this planning issue means each strategy has a defined scope and its dependencies and edge cases are assigned to follow-up work.

Written by the indexing model from the issue text.