fix(cli): Improve error messages in model info command with actionable guidance

[PredictiveManish]

Problem

The zse info <model> command currently shows generic error messages when a model fails to load (e.g., [red]Error analyzing model: {e}[/red]). This provides no actionable guidance to users, making it difficult to understand what went wrong and how to fix it.

Solution

Replaced the generic exception handler in _show_model_info_real() with specific error messages that provide:

• Clear identification of the error type
• Actionable troubleshooting steps
• Command examples for common fixes
• Context-appropriate solutions based on the error pattern

Changes Made

1. Added Specific Exception Handlers

Exception	Use Case	Guidance Provided
ImportError	Missing dependencies	Install commands for transformers/torch
ConnectionError	Network/proxy issues	Proxy configuration, internet check
ValueError	Invalid model name	Correct format examples
FileNotFoundError	Local path missing	Path verification, pull command
PermissionError	File permissions	Permission fixes

2. Pattern-Based Error Detection

The generic handler now detects common error patterns:

Pattern	Detection	Solution
Authentication	"401", "unauthorized"	zse login command
Rate Limiting	"429", "rate limit"	Wait or use cache
Out of Memory	"cuda", "out of memory"	Smaller model or CPU
Not Found	"404", "not found"	Verify model ID

3. Actionable Guidance Examples

Authentication Error:

$ zse info meta-llama/Llama-3-8B
Error analyzing model: AuthenticationError - 401 Client Error

Common issues and solutions:
  • This model requires authentication (gated model)
  • Log in with: zse login
  • Get token from: https://huggingface.co/settings/tokens

Invalid Model:

$ zse info llama-8b
Error analyzing model: ValueError - Model not found

Troubleshooting tips:
  • Check the model name spelling
  • Use format: organization/model-name (e.g., meta-llama/Llama-3-8B)
  • Or provide a local path to a model directory or .zse file

Examples of valid model specifications:
  meta-llama/Llama-3-8B
  Qwen/Qwen2.5-7B-Instruct
  ./models/my-model.zse
  /path/to/huggingface/model

[PredictiveManish]

@zyoraclub