Implement first baseline forecasting model #44

Goal

Implement the first baseline forecasting model for ARGUS time-series data.

Why

Before adding complex prediction models, ARGUS needs a simple and testable forecasting baseline. This creates a comparison point for future machine-learning or deep-learning models.

Scope

• Implement a simple baseline forecasting approach selected from research
• Use historical time-series data as input
• Keep forecasting logic independent from UI code
• Add basic evaluation metrics
• Add tests for expected behavior
• Document assumptions and limitations

Acceptance criteria

• Forecasting function can generate a simple prediction
• Model uses normalized historical time-series data
• Evaluation metric is calculated
• Forecasting logic is independent from NiceGUI
• Tests cover basic behavior
• Limitations are documented
• Existing tests still pass