Handle both lumped and split biomass function

[stelmo]

Some models (e.g. community models, older models) employ split biomass functions that will fail most of the biomass function tests. Make the testing procedure robust against them.

[exaexa]

Is there some way to manage that? (pick one of the biomasses?) Or does that need to completely modify the test suite?

[stelmo]

It requires us to identify when this happens and then dramatically change the test function unfortunately. Basically the biomass function for the older models looks something like this:

arginine + proline + ... + cysteine -> amino_acids
lipid_1 + lipid_2 -> all_lipids
...
amino_acids + all_lipids + ... -> biomass

I think they handle this in memote.py, but I forgot where...