Process issues in batches in order to process > 250

[jlord]

Right now there is a limit (01f84d4) of 250 set for how many issues this will do at once. This is because OSes can't have infinite files open/being written at once.

(If you really have lots of issues you'll probably hit the GitHub api limit before this hard limit 😁)

In the future it would nice to have this project handle this by doing the requests and writes in batches so make all limits observed 🌈