Added a species constraint for the max number of rings fused together

[alongd]

Motivation or Problem

Sometimes users know they are not interested in ring growth in their model, yet RMG spends significant time on suggesting species with complex fused rings. Using the maximum heavy atoms constraints isn't always helpful since other species with the same Mw could be relevant for the model.

Description of Changes

Added a maximumFusedRings key to the constraints dict.
A new method was implemented in Molecule along with a test.

Testing

A unit test was added.
FWIW, the feature was also tested locally by generating a model that previously gave many fused ring structures and verifying that they are absent now.

Reviewer Tips

Note that a previously non-implemented constraint was removed from the docs
I saw that this branch fails online due to a Julia installation, I think. Looks unrelated to this feature.

[JacksonBurns]

@alongd with the merging of #2608 and #2609 this PR should be good to review after a rebase.

A pre-emptive review comment, though. There are a few mentions of the maximumIsotopicAtoms constraint elsewhere in the documentation and examples. See: http://reactionmechanismgenerator.github.io/RMG-Py/search.html?q=maximumIsotopicAtoms+

Could you also remove these? And if there is an associated unit test for maximumIsotopicAtoms could you also remove that?

[JacksonBurns]

@alongd the new unit test is failing, see: https://github.com/ReactionMechanismGenerator/RMG-Py/actions/runs/8028132940/job/21932958791?pr=2606#step:11:1990

It seems like you expect benzene to count as 1 ring in a fused ring system:

        mol = Molecule(smiles="c1ccccc1")
        assert mol.get_ring_count_in_largest_fused_ring_system() == 1

Should this be so?

[JacksonBurns]

See #2606 (comment)

[alongd]

Sorry for not getting back to this PR, I'm having trouble with Julia and can't run the test locally. I'm getting:

Launching pytest with arguments moleculeTest.py::TestMolecule::test_get_ring_count_in_largest_fused_ring_system --no-header --no-summary -q in /home/alon/Code/RMG-Py/test/rmgpy/molecule

DEBUG (1764452) 
DEBUG (1764452) Debug-level logging is enabled for PyJulia.
DEBUG (1764452) PyJulia version: 0.6.1
DEBUG (1764452) pyprogramname = None
DEBUG (1764452) sys.executable = /home/alon/mambaforge/envs/rmg_env/bin/python
DEBUG (1764452) bindir = /home/alon/mambaforge/envs/rmg_env/share/julia/site/julia-1.8.5/bin
DEBUG (1764452) libjulia_path = /home/alon/mambaforge/envs/rmg_env/share/julia/site/julia-1.8.5/bin/../lib/libjulia.so.1

PyCall is not installed or built.  Run the following code in Python REPL:

    >>> import julia
    >>> julia.install()

See:
    https://pyjulia.readthedocs.io/en/latest/installation.html
            
Exit: PyCall not built
!!!!!!!!!!!!!!!!!!! _pytest.outcomes.Exit: PyCall not built !!!!!!!!!!!!!!!!!!!!

Process finished with exit code 1

Empty suite

I tried reinstalling Julia and rebuilding as suggested, but still get the same thing

[JacksonBurns]

Hmm that's vexing - could you try installing a more recent version of Julia?