feat: add S2 image processing and visualization

[rfievet]

Summary

This PR adds image processing, spectral indices, cloud masking, and visualization for Sentinel-2 data. This is the second of 3 PRs to add complete Copernicus support.

Builds on: PR #6 (Core Infrastructure + Basic S2)

What's Included

• RGB Extraction: Extract RGB composites from S2 ZIP files
• Spectral Indices: NDVI, NDWI, EVI, SAVI, NBR calculations
• Cloud Masking: Extract and apply cloud/shadow masks
• Visualization: Display S2 imagery, coverage maps, comparisons
• Memory Optimization: Bbox cropping to reduce memory usage by 99%+

New Capabilities

Users can now:

1. Download S2 data (from PR1)
2. Extract RGB composites for visualization
3. Calculate vegetation/water indices
4. Apply cloud masks for quality control
5. Create comparison plots and coverage maps
6. Crop to small areas to save memory

Usage Example

from src.data.copernicus import (
    CopernicusClient,
    extract_rgb_composite,
    calculate_ndvi,
    extract_cloud_mask,
    display_satellite_image
)

# Download S2 data
client = CopernicusClient()
s2_files = client.fetch_s2(
    bbox=[6.15, 49.11, 6.16, 49.12],
    start_date="2024-01-01",
    end_date="2024-01-31",
    max_cloud_cover=20.0
)

# Extract RGB composite with bbox cropping (saves 99%+ memory)
rgb_data = extract_rgb_composite(
    s2_files[0],
    bbox=[6.15, 49.11, 6.16, 49.12]  # Crop to save memory
)

# Calculate NDVI
ndvi = calculate_ndvi(s2_files[0])

# Extract cloud mask (Level-2A products only)
cloud_mask = extract_cloud_mask(s2_files[0])

# Visualize
display_satellite_image(rgb_data)