Ting-Chen

I also ran a test(which included `CH4`, `NH3`, and `O2`species) with all new libraries and surface families, and the model was completed.

The same as [O]O[Pt] species, the error messages were: ``` rmgpy.exceptions.UndeterminableKineticsError: (TemplateReaction(reactants=[Molecule(smiles="[O]O[O].[Pt]"), Molecule(smiles="O=[Pt]")], products=[Molecule(smiles="[O]O[Pt]"), Molecule(smiles="[O]O[Pt]")], pairs=[], family='Surface_Abstraction_vdW'), 'Kinetics could not be determined. Unable to find matching template for reaction +...

similar problems: If we search  the reactions with charged N2O show  when you click the 4th reaction  the charged N2O shows

The same `UnboundLocalError: local variable 'reverse_kinetics' referenced before assignment` error in [this reaction](https://rmg.mit.edu/database/kinetics/results/reactant1=1%20O%20u0%20p2%20c0%20%7B2,D%7D%0D%0A2%20N%20u0%20p1%20c0%20%7B1,D%7D%20%7B3,S%7D%0D%0A3%20X%20u0%20p0%20c0%20%7B2,S%7D__res=True)  The reverse reaction search was ok (It belongs to `Surface_Adsorption_Single` family)

#432 This PR had reported the duplicate surface reactions with different kinetics

Run the example of methane oxidation with new training data and the models completed: (RMG-Py branch is master) - RMG-database branch: **add_surface_training** The final model core has 60 species and...