Lihgt refactorization of 1D and 2D distributedGridSynchronizer. Now hav same interface as 3D distributedGridSynchronizer and function Synchronize is lighter.

1)Traversar writes to Boundary -- known BUG - this travarsar will write boundary condition in overlap too -- test is forcet to accept this
2)Traversar writes to Interior entities
3)Traversar writes to Allentities
4)Traversar writes its rank into AllEntities and then mesh si synchronized. Check Overlaping entities that data come from correct neibrouh
5)Traversar writes Linear function into AllEntities and then mesh is synchronized. All entities are checked that it has correc value.

Limitations:
Test run only on 9 process by code -> only regular 2D distribution of process is checked. Linear distribution is not chceked.
Test run only for double as RealType of DOF.
Only one size of global mesh is tested.

Working implementation of 3D distributed grid, but 3D traversars do not know about distributed grid, Process{Boundary/Interrior}Entities do not work correctly with distributed gird.

Partial implementation of 3D distributed grid. není funkční, potřebuju rychledostat rozpracovanou práci ze servrovny, kde nejde klima.

Long live clang for pointing this out!

Conflicts:
	src/TNL/Meshes/GridDetails/Traverser_Grid1D_impl.h
	src/TNL/Meshes/GridDetails/Traverser_Grid2D_impl.h
	src/TNL/String.cpp
	tests/CMakeLists.txt

Standard CMAKE_CXX_FLAGS are a space-delimited list, but our
CXX_TESTS_FLAGS are manually passed to CUDA_ADD_EXECUTABLE, which
expects a semicolon-delimited list. Blame cmake for such a crazy
language...

- Use math function from TNL/Math.h instead of the standard library
- Many cosmetic changes for readability

- The inherited operator= is weird, but this seems to work.
- Use TNL::pow instead of std::pow etc.

The splitting with if( std::is_same<...> ) is not perfect, the code for
all devices was always visible to the compiler, and irrelevant parts
were optimized out at a much later stage. To fix compilation for MIC and
custom data type, we need proper template specializations.

- Cleaned up unnecessary specializations of TNL::max and TNL::min
- Implemented result type promotion to fix compilation with nvcc
- Added TNL::sqrt, which will be necessary for MIC

It used to work everywhere, but the compilers on CentOS (on penny)
are weird...

Conflicts:
	README