Overlap comms compute improvements in hand op kernels, and better timing from Edison and Cori

lib/qcd/action/fermion/WilsonFermion5D.h

@@ -61,6 +61,7 @@ namespace Grid {
     public:
      INHERIT_IMPL_TYPES(Impl);
      typedef WilsonKernels<Impl> Kernels;
+     double alltime;
      double jointime;
      double commtime;
      double dslashtime;