Improve indexer statistics drain rate

Build 2009

I am aware that many more mutations will get sent to an index than actually have to be persisted/included in that index.

Currently, the "drain rate" seems to correlate with the total incoming mutations but it may be very helpful to see more realistically how many mutations are causing the indexer to do the bulk of its work and also to use that metric to observe how often the data related to that specific index is changing.