Release/15.0.0

[Gudahtt]

Note

Publishes v15.0.0 with stricter JSDoc requirements, escalates select rules to errors, updates TypeScript rules/peer deps, and bumps peer deps across all configs.

• Base (@metamask/eslint-config):
  • BREAKING: Expand jsdoc/require-jsdoc coverage (arrow functions, classes, TS enums/interfaces/types, methods, etc.).
  • BREAKING: Promote several promise/* rules from warn to error.
• TypeScript (@metamask/eslint-config-typescript):
  • BREAKING: Require typescript-eslint@>=8.39.0; switch to import-x/consistent-type-specifier-style preferring top-level type imports.
  • Disable @typescript-eslint/no-unnecessary-type-arguments, promise/valid-params, and import-x/no-duplicates; loosen naming for object literal properties; disable jsdoc/check-indentation.
  • Widen typescript peer range to all 5.x; fix issues with type-import grouping and TSDoc indentation.
• Test configs:
  • Jest: Bump peer on base to ^15.0.0; promote select jest/* rules to error; widen typescript peer to 5.x.
  • Vitest: Version aligned to 15.0.0; bump peer on base to ^15.0.0.
• Environment configs:
  • Browser, CommonJS, Node.js, Mocha: Bump versions to 15.0.0 and peer on base to ^15.0.0.
• Repo:
  • Root version to 15.0.0; update changelogs and comparison links.

^{Written by Cursor Bugbot for commit 5eaa19a. This will update automatically on new commits. Configure here.}

[Gudahtt]

I've tested this in core, and it generally seems OK except that the auto-fixer for jsdoc/require-jsdoc is awful and had to be disabled.

The auto-fixer for import-x/consistent-type-specifier-style had some issues as well, but they were manageable (when extracting inline types to be top-level, it failed to consolidate the new top-level type import with other per-existing top-level type imports).

[cryptodev-2s]

LGTM! One question though: without prior knowledge of how versioning works in this repo, it’s easy to miss the impact of a major bump, especially since we’re using workspace-based versioning (e.g. "@metamask/eslint-config": "workspace:^"). Should we consider using version sync like we do in core, so that major changes are more visible and coordinated across packages?

[Gudahtt]

Good point, I've documented that problem here: #432