docs: add error handling examples for ML inference

[addnad]

What This Fixes

Closes #228

Problem

The ML Inference documentation shows basic examples but lacks error handling guidance. This can lead to:

• Developers not knowing how to handle network failures
• Unexpected crashes from invalid model CID
• Silent wallet balance issues
• Timeouts without retry logic

Solution

Added comprehensive error handling documentation including:

• Basic try/except pattern
• Specific exception handling for common errors
• Common error reference table with solutions
• Real examples developers can copy/paste

Changes Made

• Added "Error Handling" section to ML Inference docs
• Included examples for ValueError, RuntimeError, generic Exception handling
• Added troubleshooting table for common issues

Testing

✅ All examples are syntactically valid Python
✅ Error handling patterns follow SDK conventions
✅ Documentation is clear and actionable