Fixes problem that NoneTypes are TypeErrors

[MilitaoLucas]

The HBP on the API can return None if there is disorder in the coformers. None raises a TypeError in the MHBP. NoneTypes don't fall under RuntimeExceptions even though they are technically r This could be said to be a bad practice as some errors are due to problems that aren't individual to the coformer, but general system errors, like the R one. But, RuntimeException had the same problem, so I don't think in this case this is particularly bad.
I think that maybe in the future, the API could raise an DisorderException so users can treat exceptions less generally.

[pmbulit]

What about doing this on l:269 instead? Not sure if the issue also happens with donors, but testing with FUTWOI37 this only happened for acceptors.

        try:
            ascores[a.label] = [round((coordination_scores.predictions_for_label(a.label, 'a')[1])[k], 3)
                            for k in coord_cols]
        except TypeError:
            raise RuntimeError

[MilitaoLucas]

What about doing this on l:269 instead? Not sure if the issue also happens with donors, but testing with FUTWOI37 this only happened for acceptors.

        try:
            ascores[a.label] = [round((coordination_scores.predictions_for_label(a.label, 'a')[1])[k], 3)
                            for k in coord_cols]
        except TypeError:
            raise RuntimeError

I don't really know. But I think there is not much improvement on being more specific at this point. A RuntimeException already is very general, so being more general is not that bad IMO. I also tried doing this for acceptors and donors on propensity calc but thought it would be meaningless repetition. It could work, but I think it would be better to be more general and avoid any future errors that stop the entire run. It is extremely frustrating to let a library run overnight, being excited to do some experiments in the morning and being met with an exception exclusive to one pair API-coformer that stops everything.

[pmbulit]

Fair. The idea with the exceptions is that the run doesn't fail because one calculation failed so I've approved the changes