correct OverdeterminationValidator for L3V2

[niko-rodrigue]

In L3V2, you can have inside the mathML the csymbol rateOf on top of reaction id. The current code does not take this into account and the validation is not properly detecting problems from some L3V2 models.

[dyrpsf]

Hi @niko-rodrigue & @draeger ,

I’ve opened a pull request to address this: #276

Summary of the change:

1. Updates OverdeterminationValidator so that when parsing MathML, it correctly accounts for identifiers used as targets of the L3V2 rateOf csymbol.

• If ASTNode.getVariable() is null (which can happen for rateOf(reactionId)), the validator now resolves the identifier string in the model (species, compartment, parameter, or reaction) and includes that SBase when building the bipartite graph.

2. Adds a unit test OverdeterminationValidatorTest that verifies a reaction referenced only by its id in an AST node is still collected as a variable.

This should make overdetermination detection work correctly for models using rateOf in L3V2. Feedback or additional test cases are very welcome.