[openvino-langchain] Use applyChatTemplate in ChatOpenVINO

[Retribution98]

Details:

• Use applyChatTemplate to convert messages
• Update LLMPipeline.generate to use DecodedResults and avoid a warning
• Fix row lengths

Tickets:

CVS-172173

[Copilot]

Pull Request Overview

This PR updates the OpenVINO-LangChain integration to use the applyChatTemplate method for converting messages in the ChatOpenVINO class, improving compatibility with the underlying tokenizer. The changes also address warnings and fix formatting issues.

• Replaces manual message conversion with applyChatTemplate API call
• Adds return_decoded_results flag to avoid type warnings in generation options
• Fixes line length formatting in error message

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
chat_models.ts	Implements applyChatTemplate for message conversion and adds decoded results flag
llms.ts	Adds return_decoded_results flag to generation options
embeddings.ts	Splits long error message across multiple lines for better formatting

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[apaniukov]

Why is the add_generation_prompt false?

[Retribution98]

Thanks, I fixed it.

[almilosz]

Awsome