Implement error handling - data source and parsing problem

[bobleesj]

The following is a technical comment by one of the JOSS reviewers.

Error handling for 1) .cif files that do not contain specific attributes 2) invalid input file.

---

Just a few of final minor points:

• ... (omitted)
• ... (omitted)
• I tried some more random CIF files from CSD/ICSD/MP etc. and all now work without any errors (:confetti_ball:), but some functionality that I expected to work did not. For example, in cases where the CIF only provides chemical_formula_sum and not chemical_formula_structural, the histogram generation for an formula/structure etc. for the ensemble fails. Similarly, when trying to generate polyhedra for these CIFs, the program hangs in cases where invalid input is provided (i.e., not providing a label). You could say these are either user errors or data source errors, but I think for a package aimed at non-specialists it would be helpful to provide a clear error message and fail early in cases where this info is missing. I don't see this as blocking for the paper.

Originally posted by @ml-evs in openjournals/joss-reviews#7205 (comment)