Lexical operations

Nobody has claimed this yet.

## Parse Label to OWL def

Given a string literal, and a literal-generating DP, test to see if the string parses using the DP, and if so, generate the tuples (or the OWL def, or the full class)

Return also some kind of structured evidence - e.g. if uses a 'related' uberon synonym with xref FOO:123, include this (this is useful for weighting candidate derived axiom)

## Propose lexical generator rule

Given:

* string literal
* DP name
* tuple

(or replace the last 2 with an OWL def, which can be used to derived DP+tuple)

Make a proposed lexical generating rule.

E.g. ` ("abnormal limb morphology", abnormal_morphology, UBERON:999) ==> 'abnormal %s morphology', structure`

## Propose label in referenced class

Given

* DP, with lexical generator rule
* tuple with slots S1...Sn (slots for the defined class)
* label for the defined class

propose a label for a class S_i, such that this is a new label/syn, and the lexical generator rule for the defined class can use this to generate the defined class label.

E.g. "abnormal hand morphology", has one slot, filled with uberon class 'manus'. Suggest new uberon synonym 'hand'

## Learning

TODO: make this a distinct ticket. With the above, we can learn new DPs and use to generate missing parts of ontologies

No implementation files, tests, or entry points are named. Start by narrowing the three proposed operations—parsing labels, proposing lexical generator rules, and proposing labels for referenced classes—and define their inputs, structured evidence, and acceptance criteria before implementation.

Written by the indexing model from the issue text.