Improve profiling information

[leonhard-s]

Additional stats and metrics to provide to the user to allow them to judge the health of their application. This has no set date or priority yet, just collecting ideas for now.

Rest API

• Requests per second
• Requests in the last minute
• Average latency
• Latency deviation (API likes to throw in some spikes here and there)
• Max retries in the last 5 minutes
• Query retry frequency

Cache Debug

• Cache size
• Cache usage
• Avg. time in cache
• Proportion of timeouts vs. pushouts

Object Model

• Caching ratio per datatype (i.e. how many instances go through to the API vs. are restored from cache)