Allow the user to select a model

[epylar]

• Implement argparse for handling command line options
• Add --model option to allow specifying the model name (default: gpt-4)
• Update send_error_to_gpt4 function to accept model_name as a parameter
• Rename send_error_to_gpt4 function to send_error_to_gpt

[epylar]

addresses #2, #3 and partially addresses #5

[biobootloader]

Thank you! I'd actually already reworked to use Fire for arguments, I've pushed that now. I'll fix the new conflicts and merge your other changes