fate of MHM2MRM definition

Currently, there is this line in the top-level CMakeLists.txt:

cpp_definitions("-DMRM2MHM" "CMAKE_MRM2MHM" "ON" "If set to ON the model runs with mRM")

so that we be default compile mHM with mRM. If one would turn it off, there will be errors, e.g:

Error: Symbol ‘kge_q’ at (1) has no IMPLICIT type; did you mean ‘kge_et’?
/Users/ottor/nc/Home/local_libs/fortran/mhm_mpr/src/mHM/mo_objective_function.f90:854:63:

  854 |     call init_indexarray_for_opti_data(domainMeta, 1, nQDomains, opti_domain_indices_Q)
      |                                                               1
Error: Symbol ‘nqdomains’ at (1) has no IMPLICIT type; did you mean ‘netdomains’?

because nQDomains is defined within the #def MHM2MRM block but also called outside it.

What is the policy on the MHM2MRM definition (@muellese, @thober)?

• keep it, then we should also test it and fix it (maybe @kaluza?)!
• drop it, then we can get rid of it altogether!