feat(Germany): Add Germany solar generation data pipeline

[Sharkyii]

Pull Request

Description

Adds a complete end-to-end pipeline for Germany solar PV forecasting - including data download scripts for SMARD PV generation and GFS weather data, a processing/validation pipeline, and a baseline model training script, all configurable via YAML configs

Fixes #121

How Has This Been Tested?

All scripts were validated end-to-end against real data:

PV data: 34,944 time steps, 1 GSP region (Germany 2021)
GFS data: 28 init times, 17 forecast steps, 33×41 spatial grid
Verified data loading, validation (negative value checks, NaN percentage), temporal alignment, and normalization constant calculation

• Yes

If your changes affect data processing, have you plotted any changes? i.e. have you done a quick sanity check?

• Yes

Checklist:

• My code follows OCF's coding style guidelines
• I have performed a self-review of my own code
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works
• I have checked my code and corrected any misspellings

[Sharkyii]

Adds a complete end-to-end pipeline for training solar PV forecasting models for Germany, using GFS weather data (NOAA) and PV generation data from the SMARD API.

What's included

• Data ingestion: Scripts to download PV generation data (downloading_pv_germany.py) and GFS weather forecasts (download_gfs_germany_fast.py), with support for both recent and historical data via herbie-data
• Processing pipeline: germany_pipeline.py orchestrates inspection, validation, temporal alignment, and normalization constant generation
• Model training: train_germany_baseline.py trains a baseline forecasting model from processed Zarr data
• Configs: GFS download settings, PV data config, regional boundaries, and PVNet datamodule configuration for Germany
• Utilities: Shared helpers in germany_utils.py

Data format

• PV: (datetime_gmt, gsp_id) — generation_mw, capacity_mwp
• GFS: (init_time_utc, step, lat, lon) — 14 weather channels at 0.25° resolution

Checkout - README

Next Steps

Train Model still in development