[cluster] keep track of node counts cluster-wide.

This generalizes exchange of signals between the ranks using a non-blocking all-reduce. It is now used for the stop signal and the node count, but should be easily generalizable (TB hits, and ponder still missing). It avoids having long-lived outstanding non-blocking collectives (removes an early posted Ibarrier). A bit too short a test, but not worse than before:

Score of new-r4-1t vs old-r4-1t: 459 - 401 - 1505  [0.512] 2365
Elo difference: 8.52 +/- 8.43

src/timeman.h

@@ -23,7 +23,7 @@
 
 #include "misc.h"
 #include "search.h"
-#include "thread.h"
+#include "cluster.h"
 
 /// The TimeManagement class computes the optimal time to think depending on
 /// the maximum available time, the game move number and other parameters.
@@ -34,7 +34,7 @@ public:
   TimePoint optimum() const { return optimumTime; }
   TimePoint maximum() const { return maximumTime; }
   TimePoint elapsed() const { return Search::Limits.npmsec ?
-                                     TimePoint(Threads.nodes_searched()) : now() - startTime; }
+                                     TimePoint(Cluster::nodes_searched()) : now() - startTime; }
 
   int64_t availableNodes; // When in 'nodes as time' mode