Update the test embedding model service

What does this MR do and why?

We expect the embedding model testing to always be invoked by a user, so the TestEmbeddingModelService should require a user.

Additionally, we need a way to test the /v1/embeddings/code_embeddings/search endpoint, so TestEmbeddingModelService now accepts an optional boolean search parameter. This currently is not used in production, but rather used locally to test requests to the search endpoint coming from Rails.

References

• Related to https://gitlab.com/gitlab-org/modelops/applied-ml/code-suggestions/ai-assist/-/work_items/1985+ - the TestEmbeddingModelService is used to test and verify the billing events emitted for requests coming from Rails with the right headers.

Screenshots or screen recordings

N/A

How to set up and validate locally

Test on the Rails console

Ai::ActiveContext::TestEmbeddingModelService.new(
  collection_class: ::Ai::ActiveContext::Collections::Code,
  model_ref: 'text_embedding_005_vertex',
  model_type: 'gitlab_managed',
  dimensions: 768,
  user: User.first
).execute

# test with `search=true`
Ai::ActiveContext::TestEmbeddingModelService.new(
  collection_class: ::Ai::ActiveContext::Collections::Code,
  model_ref: 'text_embedding_005_vertex',
  model_type: 'gitlab_managed',
  dimensions: 768,
  user: User.first,
  search: true
).execute

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.