Identifier types as a vocabulary instead of subclasses

[frosterus]

Would it make sense to change how the different identifier types are handled in BIBFRAME? This is related to issue #120 .

It seems cumbersome to include the various identifier types as dedicated Classes in the BIBFRAME datamodel. We at NLF have a need for a few new identifiers as well, but it seems it might make more sense to model identifiers similar to how, e.g., note types are handled using a separate vocabulary (https://id.loc.gov/vocabulary/mnotetype.html). This solution would not put stress on the data model when new identifier types are inevitably needed. The LC Standard Identifier list could work for this as is (https://id.loc.gov/vocabulary/identifiers.html).