Clappmond RAS enhancemenets

Recently I worked on a number of calls where PHA application monitor timed out. Usually we can not provide a proper RCA in these cases which frustrates our customers.
I propose to add an option to application monitor / clappmond to collect additional debug data if the application monitor times out.

1. Run pdump.sh (https://www-01.ibm.com/support/docview.wss?uid=aixtools650ae3be) on clappmond and all of its child processes - this is useful when the actual monitor script or one of its child process hang.

2. Collect perfpmr data for 2-3 minutes.

In both cases PHA should collect the debug data before killing the monitor process and performing any further action (e.g. server restart or takeover).

Additional notes:
- This could be added into the existing cl_ffdc event script, though we may have to change how and when the daemon invokes it in order to get relevant data before any recovery.
- Using the existing event should also be enough to trigger an upstream event for the SMUI so we could add notifications in smui itself.
- Running anything for 2-3 minutes before doing recovery is not something we would want as default behavior, especially in a production environment, so we would have to come up with some approach for enabling this only as needed.
- It looks like both pdump.sh and perfpmr have to be downloaded separately – it would be nicer if they were shipped with base aix, but I suppose if the customer gets to the point where they are needed, that the downloading itself should not be a concern. We would also have to integrate the collection of data from these tools with cl_ffdc.
- Add some kind of simple locking (e.g. a lock file) to avoid running multiple perfpmrs at the same time