Add generic kernel instances for ck::tensor_operation::device::DeviceGemmMultipleD

[geyyer]

Adding generic kernel instances requested in #709.

• Add a gemm_add_add_fastgelu generic kernel instance, add a client example.
• Same for gemm_add_fastgelu and gemm_fastgelu.
• Tests:

• ./02_gemm_add_add_fastgelu/client_gemm_add_add_fastgelu_generic 1 73 197 -1 -1 -1 -1 -1
  

• ./02_gemm_add_add_fastgelu/client_gemm_add_fastgelu_generic 1 73 197 -1 -1 -1 -1
  

• ./02_gemm_add_add_fastgelu/client_gemm_fastgelu_generic 1 73 197 -1 -1 -1