New acceptor molecule workflow needs a tweak to the indexing

An adaptation of this workflow

https://github.com/cmelab/morphct/blob/master/examples/new-acceptor-molecule-workflow.ipynb

worked for @chemicalfiend (@gwenwhite) on y6 and, because y6 and BTO have the same number of atoms, it should work for any mixture of these two. However, with some clever numpying, the code that automates the chomophore list building could be modified to handle any mixtures.

Start with examples/new-acceptor-molecule-workflow.ipynb and trace the indexing and chromophore-list-building steps used for Y6 and BTO. Compare the assumptions about atom counts, then determine whether the workflow handles mixtures with differing atom counts; done means the automated list-building works for arbitrary mixtures described in the issue.

Written by the indexing model from the issue text.