rename water_vapour_m_x to predicted_water... in condensation solver where needed

[slayoo]

[slayoo]

@tluettm, trying to implement this change, I've noticed that the condensation code is structured in a way that it takes a copy of the "current" moisture/theta var, modifies this copy internally, and returns it to be used at calling scope as the new "predicted" state. I'm wondering how to make the var names more readable? Renaming the argument to predicted_water_vapour_mixing_ratio will be equally misleading, as what is passed in it are the "current" values, while when it is returned, these are the new "predicted" ones...