The generated TRM project in the SaturatedPointheatsource.ipynb notebook
does not converge otherwise.

Changed some solver(preferred) or test tolerances slightly to get tests satisfied on BGR servers

Slightly change the test to see influence of deactivated temperature
field on the pressure variable.

The original input is not close enough to unity.

HoekBrown and a MFront test

Testing vectorial quantities evaluation in exprtk Function MPL property.
Testing that a Kelvin vector is converted to symmetric tensor,
deformation gradient.
The avg function is good for testing because it catches various possible
issues with the sizes of the vectors being tested.

Also adding a check for nan in evaluation.

The PIMPL removes the exprtk header from the header file.
And, what is more important, prepares ground for vectorial quantities
evaluation.

Linking to exprtk is neccessary because MaterialLib designates
private linking to that library.

The setting was provided by a user and didn't work properly. With the
previous commits it works now.

Using std::pair wasn't possible because the unit unit test for
FixedTimeStepping (implemented using autocheck) doesn't compile

Note: No reference data has been added, but existing data has been
renamed and used.

Warning was:

C:\Program Files (x86)\Microsoft Visual
Studio\2019\Community\VC\Tools\MSVC\14.29.30133\include\memory(3382):
warning C4244: 'argument': conversion from 'const uintptr_t' to 'const
unsigned int', possible loss of data
C:/Users/gitlab/gitlab/_b/bg4d5s_d/0/ogs/ogs/Tests/ProcessLib/TestReflectIPData.cpp(792):
note: see reference to function template instantiation
'std::unique_ptr<LocAsm,std::default_delete<LocAsm>>
std::make_unique<LocAsm,const size_t&,0>(const size_t &)' being compiled

This requires separation of enthalpies into FluidEnthalpyData
and SolidEnthalpyData, hence the many changes.

It seems to avoid some issues after Eigen update with the SparseLU
solver crashing.