Add MCP (Model Context Protocol) Support to UniGetUI

### Please confirm these before moving forward.

- [x] I have searched for my feature proposal and have not found a work-in-progress/duplicate/resolved/discarded issue.
- [x] This proposal is a completely new feature. If you want to suggest an improvement or an enhancement, please use [this template](https://github.com/marticliment/WingetUI/issues/new?assignees=marticliment&labels=enhancement&projects=&template=enhancement-improvement.yml&title=%5BENHANCEMENT%5D+%28Enter+your+description+here%29).

### Describe the new feature

I’d like to request the addition of Model Context Protocol (MCP) support in UniGetUI. MCP is an open standard that enables AI models, such as large language models (LLMs), to interact with external tools and systems. By implementing an MCP server within UniGetUI, the package manager could allow LLMs to manage software packages through natural language commands. For instance, users could use LLM-CLI tools like Gemini-CLI, Claude-Code, or Qwen-CLI to say, "install package X using UniGetUI," and the AI would communicate with UniGetUI via MCP to execute the action.

### Describe how this new feature could help users

Motivation
The purpose of this feature is to enhance UniGetUI’s accessibility and integration with AI-driven workflows, which are growing in popularity. MCP provides a standardized way for AI models to connect to tools like UniGetUI, eliminating the need for custom integrations. This would enable:

- Natural language package management: Users could install, update, or remove packages using voice or text-based AI assistants.
- Automation: AI tools could streamline software management tasks within larger workflows.
- Broader appeal: UniGetUI would become more versatile for users leveraging AI in software development or system administration.

Since UniGetUI already supports a CLI, MCP could extend this capability to AI contexts, making it a natural evolution of its existing features.

Possible Implementation
UniGetUI could implement an MCP server to expose its package management functions (e.g., search, install, update, remove) as tools that AI models can access. Two potential approaches include:

- CLI Wrapper: Build an MCP server that interfaces with UniGetUI’s existing CLI commands. This would translate natural language requests into CLI instructions, leveraging current functionality.
- Direct Integration: Expose UniGetUI’s internal APIs or functions directly through an MCP server for potentially better performance and control.

The MCP server would define tools corresponding to UniGetUI’s core operations, making them available to LLM-CLI tools.

Additional Considerations
- Security: Since MCP allows external AI models to execute commands, UniGetUI should include authentication (e.g., API keys) and authorization to prevent unauthorized access or misuse.
- Performance: Running an MCP server alongside UniGetUI should be optimized to avoid impacting its core performance, especially under multiple requests.
- Documentation: Provide clear instructions and examples for users to connect LLM-CLI tools (e.g., Gemini-CLI, Claude-Code, Qwen-CLI) to UniGetUI’s MCP server.

Conclusion
Adding MCP support would make UniGetUI more future-proof and appealing to users integrating AI into their workflows. It aligns with the growing adoption of MCP as a standard (supported by companies like Anthropic and GitHub) and could position UniGetUI as a forward-thinking package manager. Given its open-source nature, the community could also contribute to this feature’s development.