Description

This PR reviews the fields in the Conversation API from a Dapr Agents perspective, evaluates which ones we actually need, and aligns our API more closely with OpenAI’s where it makes sense (ie adding fields/commenting on ones we can add in future). Specifically, I’ve:

• Updated the meaning of the metadata field. This is not component metadata (as some quickstarts/docs currently imply), but rather additional metadata passed through to the LLM provider. I do need feedback on this please.
• Identified two fields we don’t use at all (context_id and parameters) and marked them with TODO comments for removal once the API is stable for in future.
• Added a response_format field, which is required for Dapr Agents when using the Conversation API to request responses in a specific JSON schema.
• Bumped the LangChain package version.

Still TODO:

• Update integration tests to reflect these changes.
• Open contrib PR bubbling these fields over to the LLM providers and fully test out these fields for the different conversation components
• I will likely open a separate PR or just add in usage metrics fields too on the response

Opening to get early feedback on the changes.

Issue reference

Please reference the issue this PR will close: #[issue number]

Checklist

Please make sure you've completed the relevant tasks for this PR, out of the following list:

• Code compiles correctly
• Created/updated tests
• Unit tests passing
• End-to-end tests passing
• Extended the documentation / Created issue in the https://github.com/dapr/docs/ repo: dapr/docs#[issue number]
• Specification has been updated / Created issue in the https://github.com/dapr/docs/ repo: dapr/docs#[issue number]
• Provided sample for the feature / Created issue in the https://github.com/dapr/docs/ repo: dapr/docs#[issue number]