Add image recognition tools

[NakaokaRei]

Description

Add tools for finding images on screen using computer vision.

SwiftAutoGUI API Reference

• locateOnScreen(imagePath: String, confidence?: Float) - Find image on screen
• locateCenterOnScreen(imagePath: String) - Find image center
• locateAllOnScreen(imagePath: String) - Find all image matches

Proposed Implementation

locateOnScreen

• Input:
  • imagePath (string) - Path to reference image
  • confidence (number, optional) - Match confidence 0.0-1.0
• Output: {x, y, width, height} or null if not found

locateCenterOnScreen

• Input:
  • imagePath (string) - Path to reference image
  • confidence (number, optional) - Match confidence
• Output: {x, y} center coordinates or null

locateAllOnScreen

• Input:
  • imagePath (string) - Path to reference image
  • confidence (number, optional) - Match confidence
• Output: Array of {x, y, width, height} for all matches

Use Cases

• Automated UI testing
• Waiting for UI elements
• Game automation
• Visual verification

Technical Considerations

• OpenCV dependency (already in SwiftAutoGUI)
• Performance impact of image matching
• Handling retina displays and scaling
• Image format support
• Error handling for missing images