phsl / new-api

- Replace ThinkingAdapterMaxTokens with a more flexible DefaultMaxTokens map
- Add support for model-specific default max tokens configuration
- Update relay and web interface to use the new configuration approach
- Implement a fallback mechanism for default max tokens

- Update Claude relay to set default MaxTokens dynamically
- Modify web interface to clarify default MaxTokens input purpose
- Improve token configuration logic for thinking adapter models

- Introduce a new configuration management approach for model-specific settings
- Update Gemini settings to use the new config system with more flexible management
- Add support for dynamic configuration updates in option handling
- Modify Claude and Vertex adaptors to use new configuration methods
- Enhance web interface to support namespaced configuration keys

- Enhance channel test to detect more embedding models
- Update Azure OpenAI default API version to 2024-12-01-preview
- Remove redundant default API version setting in channel edit
- Add user cache writing in channel test

- Add new context keys for user-related information
- Modify user cache and authentication middleware to populate context
- Refactor quota and notification services to use context-based user data
- Remove redundant database queries by leveraging context information
- Update various components to use new context-based user retrieval methods

feat: Add Claude 3.7 Sonnet thinking mode support

# Conflicts:
#	relay/channel/claude/dto.go

feat: Pass extra_body in OpenAI request to the backend

feat: Add model request rate limiting functionality