Add CLI, batch processing, and professional tooling to Fruitify

[Copilot]

Transforms basic prediction scripts into production-ready CLI tools with batch processing capabilities.

Changes

CLI Interface

• main.py & top3.py: Added argparse for flexible image paths, custom models/labels, and configurable top-N predictions
• All scripts include --help with usage examples
• Proper error handling with stderr output and exit codes

Batch Processing

• batch_predict.py: Process multiple images with CSV export
• Directory scanning with configurable sample filtering (--include-samples)
• Progress tracking and per-image error reporting

Code Quality

• Docstrings and type hints across all functions
• Context managers for file handling
• Documented compile=False for TensorFlow/Keras version compatibility
• Modular functions for reusability

Documentation

• requirements.txt: tensorflow, keras, pillow, numpy
• CONTRIBUTING.md: Contribution workflow and code style guidelines
• CODE_OF_CONDUCT.md: Community standards
• example.py: Programmatic usage patterns
• Updated README with CLI examples

Usage

# Single prediction
python main.py fruit.jpg

# Top 5 predictions
python top3.py --top 5 fruit.jpg

# Batch processing
python batch_predict.py --input ./images --output results.csv

Maintains backward compatibility with programmatic usage.

Security

✅ CodeQL: 0 vulnerabilities

Original prompt

look for any enhancement we can apply to this project

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.