Make harvester implementation simpler/more reliable

The harvester extension has good interfaces for viewing job progress and submitting jobs, but implementing reliable harvesting jobs is tricky and when a job stops it can be hard to debug the cause.

Can we fix or replace the current harvester back end to make it easier to extend and debug?

The issue names the harvester extension and its backend but does not identify files, tests, or entry points. Start by locating the harvester implementation and reviewing how job progress, submission, stopping, and failure reporting work. Done means the backend is easier to extend and the cause of stopped jobs can be diagnosed reliably.

Written by the indexing model from the issue text.