gemv: call scal for degenerate matrices

[ndellingwood]

Related to #539

[ndellingwood]

@kyungjoo-kim I made the change you suggested, quick Cuda test with cublas and blas tpls now show gemv tests pass for degenerate cases

[ndellin@kokkos-dev-2 cmTestAllSandia]$ cd kokkos-kernels/testing/cmTestAllSandia

[ndellin@kokkos-dev-2 cmTestAllSandia]$ ../../scripts/cm_test_all_sandia cuda --spot-check --with-tpls=cublas,blas
Running on machine: kokkos-dev-2
WARNING!! THE FOLLOWING CHANGES ARE UNCOMMITTED!! :
 M src/blas/KokkosBlas2_gemv.hpp

Repository Status:  c80785a6ea35f35e460a22cb9ef35789fd1871a4 Merge pull request #566 from kyungjoo-kim/fix-cmake-fortran-mangle2


Going to test compilers:  cuda/10.1
Testing compiler cuda/10.1
  Starting job cuda-10.1-Cuda_OpenMP-release
kokkos options: 
kokkos devices: Cuda,OpenMP
kokkos cxx: -O3 -Wall -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized 
  PASSED cuda-10.1-Cuda_OpenMP-release
#######################################################
PASSED TESTS
#######################################################
cuda-10.1-Cuda_OpenMP-release build_time=164 run_time=134

[ndellingwood]

I'll run full spot-check to make sure nothing was unexpectedly disrupted, just wanted to document that tests are now passing with tpls enabled before I go home.

I'm updating the cm_test_all_sandia script to test with tpls enabled with some builds during spot-check, that PR will follow this one.

[ndellingwood]

kokkos-dev-2 results (without tpls enabled)

Running on machine: kokkos-dev-2
WARNING!! THE FOLLOWING CHANGES ARE UNCOMMITTED!! 

Repository Status:  881b411e778355560f9f05d5a053b2518f0c290d gemv: call scal for degenerate matrices

Going to test compilers:  gcc/7.3.0 gcc/8.3.0 gcc/9.1 intel/18.0.5 clang/8.0 cuda/10.1

...

#######################################################
PASSED TESTS
#######################################################
clang-8.0-Cuda_OpenMP-release build_time=116 run_time=152
clang-8.0-Pthread_Serial-release build_time=49 run_time=142
cuda-10.1-Cuda_OpenMP-release build_time=165 run_time=140
gcc-7.3.0-OpenMP-release build_time=46 run_time=47
gcc-7.3.0-Pthread-release build_time=43 run_time=74
gcc-8.3.0-Serial-release build_time=43 run_time=71
gcc-9.1-OpenMP-release build_time=52 run_time=47
gcc-9.1-Serial-release build_time=45 run_time=71
intel-18.0.5-OpenMP-release build_time=130 run_time=46