Realign Biolink Model `Attribute` to final TRAPI standard

[RichardBruskiewich]

See https://github.com/NCATSTranslator/ReasonerAPI/issues/192 and associated PR https://github.com/NCATSTranslator/ReasonerAPI/pull/186.

The final TRAPI schema nomenclature and content of their Attribute schema should be closely reflected in the Biolink Model

[nlharris]

is this still needed? is the TRAPI schema now final?

[sierra-moxon]

@RichardBruskiewich - is this still needed? In my mind, all sorts of biolink properties can be in the TRAPI message as an attribute_type_id and (sometimes) value?