Fix false positives when validating parameter table

[dweindl]

For single-row validation, don't pass any other dataframes/models to petab.check_parameter_df to avoid false positives due to the incomplete parameter table.

Closes #126.

[PaulJonasJost]

Wanted to experiment with this as well. I guess this should be done across the different tables as well.

The main question is here whether we should catch the false positives but have more check for a single row or says that a single row can only be checked without the context.

Would propose to implement a check_full setting to enable checking the full PEtab even during edit (for smaller problems). But this is a concern for another PR.

[dweindl]

Wanted to experiment with this as well. I guess this should be done across the different tables as well.

I didn't run into false positives there (yet). If there are such situations, then yes.

The main question is here whether we should catch the false positives but have more check for a single row or says that a single row can only be checked without the context.

Would propose to implement a check_full setting to enable checking the full PEtab even during edit (for smaller problems). But this is a concern for another PR.

I'd leave the rest as is for now. We can think about how to best address this in the petab library to make things easier here.

[PaulJonasJost]

Merging and Opening a small discussion (no need to contribute to it yet)