feat: Release v1.8.0 - Context Management and Dependency Refactor

- Context: Added per-provider 'context_limit' and automatic history pruning/distillation to prevent failover loops.
- Config: Added '/config budget' command and '/config set ... context_limit' support.
- Fix: Resolved critical code duplication in OpenAICompatibleProvider causing double responses.
- Fix: Fixed UI buffer duplication during retry/failover events.
- Refactor: Consolidated dependencies into pyproject.toml, simplifying requirements.txt and CI/CD.
- Tests: Converted verification scripts to pytest and achieved 100% pass rate.
- Docs: Updated documentation for context management and new commands.

Release v1.6.0 - Autonomous Mode, Hardened Sandbox, and SDK Migration

feat: implement hybrid MCP configuration support

- Added MCPServerConfig for Claude Desktop-style stdio MCP configuration (command, args, env).
- Updated MCPClient to support explicit execution arguments and environment variable injection.
- Implemented legacy MCP provider compatibility by parsing stdio:// URLs into MCPServerConfig.
- Unified MCP lifecycle management in Router.reload_providers for both config styles.

docs: update security policy to latest-only support model

feat: multi-line input and robust provider validation improvements

- Replaced single-line Input with multi-line TextArea and added asynchronous message queuing.
- Added Ctrl+Enter support for submission and refined input UI styling.
- Resolved Gemini '400' and OpenAI '500' role validation errors by mapping 'system' turns to 'user'.
- Fixed provider compatibility with local models by merging assistant messages and stabilizing tool IDs.
- Improved tool result extraction and cleaned up unused imports.