with template build configuration to speed-up the build process.

Adding solver monitor to the navier-stokes example.

Implementing tnl-diff.

Fixing the navier-stokes example.

Fixing the simple solver example.

Explicit instantiation of template CUDA reduction.

TNL can be compiled without CXX11 support.

Adding explicit template instantiation.
Moving some _impl.h files into implementation dir.

Adding directory legacy for old code.

Implementing PDE solvers and time steppers.

Changing explicit solvers - Merson and Euler.