Learning backend cuda

Context

This MR is intended to add the forward implementations for Accuracy computation operators:

• And
• ArgMax
• ReduceSum

Modified files

• AndImpl.hpp, AndImpl_forward_kernels.hpp, AndImpl.cpp and Tests_AndImpl.cpp, add And forward impl;
• ArgMaxImpl.hpp, ArgMaxImpl_forward_kernels.hpp, ArgMaxImpl.cpp and Tests_ArgMaxImpl.cpp, add ArgMaxforward impl;
• ReduceSumImpl.hpp, ReduceSumImpl_forward_kernels.hpp, ReduceSumImpl.cpp and Tests_ReduceSumImpl.cpp, add ReduceSum forward impl;

TODO

• And
• ArgMax
• ReduceSum