The previous patch made the incorrect assumption matrices were row-major
instead of column-major (got the two mixed up in my head).
This fixes this problem.
Also, this extends the test files to cover all matrix sizes (not types,
that would be redundant).
This is added as part of the GLM_EXT_matrix_common extension, as this
function is not provided by the core GLSL specification (version 4.20).
The implementation of this glm::abs(mat<>) function mirrors the
glm::abs(vec<>) implementation. It should be functionning the same in
every way as the vec implementation.
Bonus points : AFAICT this allows to vectorize operations on compilers
that support optimization of these patterns, just like the functor1<>
struct in _vectorize.hpp for vectors.
