Lenient RIS parser

We commonly see the DO - (the DOI) contain a full URL to https?://doi.org or using the doi: prefix instead of simply the DOI value.

This causes the validation to fail since we expect @Pattern(message = "Invalid DOI", regexp = "^10\\.\\d+/.+$") and the file cannot be easily imported to Genesys.

This issue is to make the parser a bit more forgiving and use only the DOI part from the DO field by extracting it using a regexp.