This change contains double-precision and single-precision optimizati…

…ons of copy routine, cpy2d-pair.

1) The copy routine cpy2d-pair is optimized to use 256-bit SIMD. The optimized version is present in the source file kernel/cpy2d-pair.c.
2) This optimization is active only for double and single data type enabled by AMD_OPT_IN_PLACE_1D_CPY2D_STABLE_INTRIN macro.
3) AMD copyright header information is added.
4) Made changes to incorporate review changes in optimized cpy2d-pair function.

Change-Id: Iac4956db0d434d8a05f97f7efb54de15b32679c6