feat(llm): add automatic provider inference for LangChain LLMs

[Pouyanpi]

Summary

Add automatic provider name detection from LLM module paths to eliminate manual provider specification. The implementation extracts provider names from LangChain package naming conventions (e.g.,langchain_openai → openai) and handles edge cases including community packages, wrapped classes, and multiple inheritance through MRO traversal.

• Add _infer_provider_from_module() to extract provider names from LangChain package conventions
• Add public get_llm_provider() API for retrieving provider names
• Update _setup_llm_call_info() to automatically infer provider when not explicitly provided
• Add comprehensive test suite covering standard packages, community packages, wrapped classes, and inheritance patterns

Test Plan

• All 9 new tests pass covering various scenarios:
  • Standard providers (OpenAI, Anthropic, NVIDIA AI Endpoints)
  • Community packages (Ollama)
  • Unknown/custom packages
  • Patched/wrapped classes
  • Multiple inheritance patterns
  • Deep inheritance hierarchies
• Existing tests continue to pass
• Provider name is correctly inferred when model_provider parameter is None

[codecov-commenter]

Codecov Report

❌ Patch coverage is 80.76923% with 5 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
nemoguardrails/actions/llm/utils.py	80.76%	5 Missing ⚠️

📢 Thoughts on this report? Let us know!

[Pouyanpi]

@greptileai

[greptile-apps]

_{2 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[Copilot]

Pull Request Overview

This PR adds automatic provider inference for LangChain LLMs by extracting provider names from module paths, eliminating the need for manual provider specification. The implementation follows LangChain's package naming conventions (e.g., langchain_openai → openai) and handles edge cases through MRO traversal.

Key Changes:

• Added _infer_provider_from_module() to extract provider names from LangChain package naming patterns
• Added public get_llm_provider() API for external provider name retrieval
• Updated _setup_llm_call_info() to automatically infer provider when model_provider is None

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
nemoguardrails/actions/llm/utils.py	Implements provider inference logic and integrates it into the LLM call setup flow
tests/test_actions_llm_utils.py	Comprehensive test suite covering standard providers, community packages, wrapped classes, and inheritance patterns

---

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

[trebedea]

Looks good!