Use consistent hash to heartbeat to a few nodes instead of all, see #2284

* Previously heartbeat messages was sent to all other members, i.e.
  each member was monitored by all other members in the cluster.
* This was the number one know scalability bottleneck, due to the
  number of interconnections.
* Limit sending of heartbeats to a few (5) members. Select and
  re-balance with consistent hashing algorithm when new members
  are added or removed.
* Send a few EndHeartbeat when ending send of Heartbeat messages.

akka-cluster/src/test/scala/akka/cluster/ClusterConfigSpec.scala

@@ -29,6 +29,8 @@ class ClusterConfigSpec extends AkkaSpec {
       PeriodicTasksInitialDelay must be(1 seconds)
       GossipInterval must be(1 second)
       HeartbeatInterval must be(1 second)
+      NumberOfEndHeartbeats must be(4)
+      MonitoredByNrOfMembers must be(5)
       LeaderActionsInterval must be(1 second)
       UnreachableNodesReaperInterval must be(1 second)
       PublishStatsInterval must be(10 second)