Add extra sleeps (conditionally) if user-ndb-object still exists during shutdown
  Note: This is not in anyway optimal, but i dont't really know in which order should be shutdown
        but it fixes problem...


sql/ha_ndbcluster_binlog.cc:
  Add extra wait loop to make user "user" ndb-object go away...
    otherwise user thread can have ongoing SUB_DATA
storage/ndb/include/ndbapi/ndb_cluster_connection.hpp:
  Add interface for getting active ndb-object count
storage/ndb/src/ndbapi/Ndbif.cpp:
  Add return in case of incorrect magic
storage/ndb/src/ndbapi/TransporterFacade.cpp:
  Add interface for getting active ndb-object count
storage/ndb/src/ndbapi/ndb_cluster_connection.cpp:
  Add interface for getting active ndb-object count