test(kernel-language-model-service): Add queue language model service for testing

[grypez]

This PR adds a queue language model service to the monorepo. The implementation is intended for mocking language model behavior where a remote API or local model execution are both unavailable. The intended streaming response is pushed to the model prior to sampling.

Also adds tests to kernel-test which exercise the basic shape of a LanguageModelService.

---

Note

Adds a queue-based LanguageModelService testing utility and integrates it into kernel tests via new vats, plus updates exports and coverage thresholds.

• Kernel Language Model Service (packages/kernel-language-model-service):
  • Test Utils: Add queue-based model/service (src/test-utils/queue/*) with makeQueueModel, makeQueueService, whitespaceTokenizer, objectResponseFormatter, and streaming helpers (normalizeToAsyncIterable, mapAsyncIterable, makeAbortableAsyncIterable, makeEmptyStreamWithAbort).
  • Exports: Expose ./test-utils entry in package.json and re-export from src/test-utils/index.ts.
  • Tests/Docs: Add unit tests for model, tokenizer, response, utils, and a README.
• Kernel Test (packages/kernel-test):
  • New Vats: Add vats/lms-queue-vat.js (envatted queue service) and vats/lms-user-vat.js (consumer).
  • Integration Test: Add src/lms-user.test.ts asserting streamed response logging; add dependency on @ocap/kernel-language-model-service.
• Tooling:
  • Coverage: Update vitest.config.ts coverage thresholds for packages/kernel-language-model-service/**.

^{Written by Cursor Bugbot for commit 9277005. This will update automatically on new commits. Configure here.}

[socket-security]

No dependency changes detected. Learn more about Socket for GitHub.

👍 No dependency changes detected in pull request

[rekmarks]

A couple of nits.

[rekmarks]

LGTM!