Suspicious algorithm when dbCa clears channels

Looking at the function addAction() in dbCa.c, it implements a behavior that looks pretty odd to me.

If it wants to add a CLEAR_CHANNEL action and there are more than <warning limit of 10000> CLEAR_CHANNEL actions still in the work list of the low level worker thread, addAction() prints a message and pauses for a hardcoded 1.0 seconds.

This doesn't seem to be the best way to handle this and doesn't scale too well.

We saw this when IOCs with many records that have a CP link to simulation switch shut down.