- Introduced LLM Provider declarations in plugin manifests, allowing plugins to specify their LLM capabilities.
- Implemented validation for LLM Provider declarations to prevent duplicates and conflicts.
- Enhanced the PluginRunner to handle LLM Provider invocation requests, enabling plugins to interact with LLM Providers seamlessly.
- Added a ClientRegistry to manage LLM Provider registrations and ensure no conflicts arise between different plugins.
- Created a PluginLLMClient to facilitate communication with LLM Providers through the plugin runtime.
- Developed tests to ensure proper registration and conflict handling of LLM Providers.
- Refactored model fetching logic to support various authentication methods for OpenAI-compatible APIs.
- Introduced new data models for LLM service requests and responses to standardize interactions across layers.
- Added an adapter base class for unified request execution across different providers.
- Implemented utility functions for building OpenAI-compatible client configurations and request overrides.
make LLM task config resolution dynamic in LLMRequest
load model clients on demand from latest config
clear client instance cache on config reload
remove stale module-level model_config usage in llm_api
add hot-reload tests for LLM/config watcher flow
