Add new vision processing nodes

[Konstantysz]

Summary

This PR introduces seven new computer vision nodes to expand the image processing capabilities of Vision-Craft, addressing issues #15, #16, #17, #18, #19, and #25.

Changes

Filter Nodes

• SobelNode: Implements Sobel edge detection with configurable derivative orders and kernel sizes
• MedianBlurNode: Provides non-linear filtering for salt-and-pepper noise reduction
• MorphologyNode: Supports seven morphological operations (Erode, Dilate, Open, Close, Gradient, TopHat, BlackHat)

Transformation Nodes

• CvtColorNode: Enables color space conversions (RGB, BGR, Gray, HSV, etc.)
• ResizeNode: Resizes images with multiple interpolation methods (Linear, Nearest, Cubic, Lanczos4)

Channel Operations

• SplitChannelsNode: Splits multi-channel images into individual channel outputs
• MergeChannelsNode: Combines individual channels into a single multi-channel image

Testing

• Added comprehensive unit tests in TestFilterNodes.cpp and TestConversionNodes.cpp
• All 256 tests pass successfully
• Verified proper integration with existing node system

Technical Details

• All nodes follow the established Nodes::Node architecture
• Proper error handling and input validation implemented
• Integrated with NodeFactory for runtime node creation
• Updated build configuration to include new source files

Future Improvements

• Add UI dropdown widgets for enum-based properties (currently using integer inputs)
• Consider adding visual property editors for more intuitive parameter adjustment

[github-actions]

📊 Test Coverage Report

✅ Coverage report generated successfully

Coverage Statistics

Metric	Coverage
Lines	82.16%
Functions	87.54%
Regions	90.09%

📁 Full HTML Report