Honor dimensions param

[jnu]

Honors dimension param in embeddings client

[sentry]

Bug: The embed_async method in OpenAIEmbeddingGenerator does not pass the dimensions parameter to the OpenAI API, unlike its synchronous counterpart embed.
_{Severity: MEDIUM}

Suggested Fix

Update the embed_async method to construct a params dictionary and pass the dimensions parameter if it's set in self.config, mirroring the logic in the synchronous embed method. Also, add a test case to verify the behavior of embed_async with the dimensions parameter.

Prompt for AI Agent

Review the code at the location below. A potential bug has been identified by an AI
agent. Verify if this is a real issue. If it is, propose a fix; if not, explain why it's
not valid.

Location: bc2/lib/embedding/openai.py#L111-L116

Potential issue: The pull request updated the synchronous `embed` method to correctly
pass the `dimensions` parameter from `self.config` to the OpenAI API. However, the
asynchronous `embed_async` method was not updated. It still calls
`self.aclient.embeddings.create` without including the `dimensions` parameter. This will
cause users who configure a specific embedding dimension and use the async path to
receive embeddings of the model's default size (e.g., 3072) instead of the requested
size (e.g., 256), leading to a silent inconsistency and potential downstream errors.

_{Did we get this right? 👍 / 👎 to inform future reviews.}