Dashboard/monitoring to surface problems

[shlomi-nx]

Dashboard/monitoring to surface problems running procedures, resolving intents

We started using Jeager, which may be sufficient. Let's see if we can get all we want. We want running statistics on:

1. Success/failure rates of procedure invocations. Perhaps break-down by failures:
   1. e.g. Find a significant increase in failures of type "no free machines" on a particular procedure
2. Time-to-completion of procedures, to see if something which takes N to complete suddenly runs significantly different to N
3. Time-to-completion of SFS - same as procedures
4. Stats on outcome of different SFS:
   1. success - successfully resolved the diff
   2. errors - Exited with some exception
   3. unresolved - no exceptions but diff still outstanding