feat: Enhance Git and AI Error Handling with Improved Diagnostics

[thomas-vilte]

Overview

I implemented a significant enhancement to the application's error handling and diagnostic capabilities, focusing on providing clearer, more actionable feedback to users. This PR addresses common pain points related to Git configuration and AI provider interactions, making the tool more robust and user-friendly.

Key Changes

• Comprehensive Error Definitions: I introduced a new set of granular error types for Git configuration (ErrGitUserNotConfigured, ErrGitEmailNotConfigured, ErrNotInGitRepo), AI provider issues (ErrGeminiAPIKeyInvalid, ErrGeminiQuotaExceeded), and VCS-specific problems (ErrGitHubTokenInvalid, ErrGitHubInsufficientPerms, ErrGitHubRateLimit).
• Git Service Enhancements: I updated the GitService to perform explicit validation of Git user configuration (user.name, user.email) before committing, and to return specific custom errors when these are missing. New methods (ValidateGitConfig, GetGitUserName, GetGitUserEmail) were added to support these checks.
• Improved Diagnostics (doctor command): I extended the doctor command to include checks for Git user.name and user.email configuration, providing immediate feedback to users about potential setup issues.
• User-Friendly Error Reporting: I significantly enhanced the HandleAppError function in the UI layer to map these new error types to specific, localized, and actionable messages and suggestions, guiding users on how to resolve issues.
• AI Output Filtering: I implemented logic within the Gemini summarizer to filter out "thinking parts" from the AI's response, ensuring cleaner and more direct output for commit suggestions.
• Internationalization Updates: I added new translation keys across English and Spanish locales to support the new error messages and diagnostic checks.

Technical Impact

• Increased Robustness: The application is now more resilient to common configuration errors, especially those related to Git setup, preventing cryptic failures during critical operations like committing.
• Enhanced User Experience: Users will receive clearer, more specific error messages and practical suggestions, reducing frustration and improving self-service troubleshooting.
• Improved AI Output Quality: Filtering AI "thought" processes leads to more concise and relevant commit suggestions, enhancing the core feature's value.
• Better Maintainability: Centralized and categorized error definitions improve code clarity and make future error handling extensions more straightforward.