Fix DotBasedGEMM bug encountered in Trilinos issue 6418

[seheracer]

Description:
This PR matches the Trilinos PR trilinos/Trilinos#6427.

Testing:

[sacer@kokkos-dev-2 unit_test]$ ./KokkosKernels_UnitTest_Cuda --gtest_filter=cuda.gemm_*
Kokkos::OpenMP::initialize WARNING: OMP_PROC_BIND environment variable not set
  In general, for best performance with OpenMP 4.0 or better set OMP_PROC_BIND=spread and OMP_PLACES=threads
  For best performance with OpenMP 3.1 set OMP_PROC_BIND=true
  For unit testing set OMP_PROC_BIND=false
Note: Google Test filter = cuda.gemm_*
[==========] Running 2 tests from 1 test case.
[----------] Global test environment set-up.
[----------] 2 tests from cuda
[ RUN      ] cuda.gemm_double
[       OK ] cuda.gemm_double (3865 ms)
[ RUN      ] cuda.gemm_complex_double
[       OK ] cuda.gemm_complex_double (3729 ms)
[----------] 2 tests from cuda (7594 ms total)

[----------] Global test environment tear-down
[==========] 2 tests from 1 test case ran. (7594 ms total)
[  PASSED  ] 2 tests.

[ndellingwood]

Whoo-hoo! Thanks @seheracer !