Add not_notable to CurationRelevance vocabulary for semi-automated …

…curation workflow (#1236)

This pull request adds the `not_notable` tag to the CurationRelevance
vocabulary as a way to mark papers that are relevant for machine
learning training but do not meet the threshold for inclusion in the
Bioregistry.

While curating papers, there have been a few instances of entries that
provide new identifier information but aren't notable enough, or
well-maintained enough for inclusion in the bioregistry
(#1225). Rather than
curating these as subpar prefixes, tagging them as `not_notable` allows
us to retain them as positive training samples without cluttering the
bioregistry with less impactful entries.

Co-authored-by: Mufaddal Naguthanawala <[email protected]>

src/bioregistry/curation/literature.py

@@ -37,3 +37,5 @@ class CurationRelevance(str, enum.Enum):
     unclear = enum.auto()
     #: Completely unrelated information
     irrelevant_other = enum.auto()
+    #: Relevant for training purposes, but not curated in Bioregistry due to poor/unknown quality
+    not_notable = enum.auto()