Allow auto-failover of nodes in server group when they fail in cascading fashion

Consider following scenario:

• Cluster has 3 or more server groups.
• Auto-failover max count = Y. 
• Server group auto-failover is enabled.
• Server GroupA has nodes Node1, Node2 … NodeN.
• Y (max count) < N (number of nodes in GroupA).
• Nodes in GroupA, fail in cascading fashion. That is first Node1 fails, is auto-failed over, then Node2 fails and so on. Eventually, all nodes in GroupA fail.

 Currently, in the above scenario, we will auto-failover Node1 … NodeY but not rest of the nodes  because the max count has already been reached.

Recently, Perry & Anil had a suggestion to allow failover of rest of the nodes because it is safe to do so.

This improvement needs non-trivial code changes and will be considered post-Vulcan.