Centralise logic for converting verb form endings and noun singular vs plural endings

There is logic scattered through maybe half a dozen linters that does things like converting plural to singular or past tense to present tense.

We have code in place for irregular verbs and irregular plurals that is already centralized.

But for regular verbs and regular nouns the logic is scattered around the linters and implemented in a variety of ways.

This is useful because regular nouns and verbs have a few patterns:
- +s
- +es
- -y +ies
- -fe +ves (nouns only)
- +double the last consonant +es

No files or tests are named. Start by locating the linters that convert regular verb and noun forms, then compare them with the existing centralized irregular-form logic; done means the listed regular patterns are handled centrally and the affected linters use that shared behavior.

Written by the indexing model from the issue text.