fix(docs): add AI limitations disclaimer and expand community warning
Created by: ihsraham
Summary
Follow-up to PR #87. @gatteo requested two additional changes to the OpenClaw community skill page:
Changes
- Expanded warning callout - Added AI risk and responsibility language to the existing community skill disclaimer at the top of the page
- New section: "Inherent Limitations of Artificial Intelligence" - Lists five key risks of AI/ML technologies when using OpenClaw, including hallucinations, biases, input errors, unreliable action suggestions, and unexpected actions (e.g. sending assets)