Allow cosolvents to be defined in a solvent library

[alongd]

Motivation or Problem

We need to define a mixture of solvents in the solvent library

Description of Changes

We now parse both a list of descriptors or just a descriptor as a solvent molecule. The descriptors could be either SMILES or adjLists.
A new method, get_structure was created in SolventLibrary

Testing

Could be tested using -db PR ReactionMechanismGenerator/RMG-database#318

[codecov]

Codecov Report

Merging #1558 into master will decrease coverage by 0.03%.
The diff coverage is 50%.

@@            Coverage Diff             @@
##           master    #1558      +/-   ##
==========================================
- Coverage   41.61%   41.58%   -0.04%     
==========================================
  Files         176      176              
  Lines       29138    29147       +9     
  Branches     5991     5995       +4     
==========================================
- Hits        12127    12120       -7     
- Misses      16172    16187      +15     
- Partials      839      840       +1

Impacted Files	Coverage Δ
rmgpy/species.py	0% <0%> (ø)	⬆️
rmgpy/rmg/main.py	22.15% <100%> (+0.06%)	⬆️
rmgpy/data/solvation.py	64.46% <100%> (-0.25%)	⬇️
rmgpy/data/kinetics/family.py	52.74% <0%> (-0.24%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 69b3b4b...cd9d91a. Read the comment docs.

[alongd]

Force pushed a small correction to the test. Should be ready for review.
@oscarwumit, could you change the corresponding database PR to have list representations so we could check this branch?

[mjohnson541]

Mostly just some comments on the new function you added. I think we should add a test demonstrating the loading of cosolvents before merging this.

[mjohnson541]

I understand the desire to reduce code duplication, but this really shouldn't be a bound function of SolventLibrary. I suggest adding a similar function to species.py or just having the code duplication.

[oscarwumit]

Force pushed a small correction to the test. Should be ready for review.
@oscarwumit, could you change the corresponding database PR to have list representations so we could check this branch?

I updated the molecule representation of co-solvent to list.

[alongd]

Thanks, @mjohnson541, addressed the comment and added a test.
(reorganized commits and force pushed)

[mliu49]

It seems like this should be called set_structure or set_molecule?

[mliu49]

I would name this variable something like spcs or spc_list for consistency with expectations.

[alongd]

Thanks @mliu49 , I addressed the comments in fixup commits

[mliu49]

Thanks! You can squash and rebase. This PR looks ok to me code-wise, though I'm not fully aware of the context for this change. It seems like @mjohnson541's comments have been resolved, so I think it could be merged if you think it's ready.

[alongd]

Thanks! Squashed and pushed.

The context is that we'd like to define a binary solvent mixture, e.g. 50% water and 50% methanol. We can get Abraham parameters from the mixture from literature (ReactionMechanismGenerator/RMG-database#318), but the solvent library forced us to define only one molecular structure for the solvent. Merging this PR will allow us to define something like:

entry(
    index = 27,
    label = "methanol_50_water_50",
    molecule = ["CO", "O"],
    solvent = SolventData(
    ...

Comments have been addressed