Normalize DOI columns for all publication files

[jacobthill]

Publications from all sources (ORCID, OpenAlex, SUL-Pub) are re-harvested in Dimensions using the publication DOI. The query will match even if the case is different, but then the output DOI that is written to file may end up having a different case than the source DOI. Also some DOIs have the web address prefix. In order to compare coverage across APIs we need all DOIs to be normalized.

To do in enrich_publication_data.py:

• .lower() all DOIs
• .replace('https://doi.org/', '') all DOIs
• Add test to ensure that all DOIs are normalized.