Most of this was related to use of BigQuery for post versions and
generating reports and bulk reverts, but elasticsearch can easily
handle returning 5k results in a few ms, so this isn't a big deal.
There is also some pubsub stuff which has been gone for a while now.
I don't remember what it was being used for, but it is likely reporting
items, so not an immediate priority.
This somewhat gimps the current source information fetching tool,
however it was already a bit suspect, and needs to be reworked before
being released as a public tool because it relies on API calls.
Makes index job unique to account for changes to update triggering.
Fixes instances of mixed SQL and ES searches with tag_matches.
Adds utility methods for exists? and count_only ES searches.
Adds comment_bumped_at for comments page usage.
Adds random seeding capability.
Fixes random searches not working correctly.
Fixes status:active.
