[Docs]: Expand COMPARISON.md to include LangChain and AutoGen

[rosspeili]

What needs to be fixed?

Our current COMPARISON.md document contrasts the Skillware philosophy strictly against the Anthropic Model Context Protocol (MCP).

Because the AI agent ecosystem is vast, we need to expand this to encompass other major frameworks to properly define our unique value proposition.

Specifically, we need to add:

1. A section detailing how Skillware's code-first, model-agnostic approach compares to standard LangChain Tools.
2. A section contrasting our "installable capabilities" with AutoGen's internal skill management.
3. Highlight our focus on uncoupled runtime logic, cognition maps, and system governance as distinct from simple execution wrappers.

Affected Page

comparison.md