feat: add EIA data preprocessing pipeline for US solar generation

[mahendra-918]

Description

Adds preprocessing for US solar data from EIA to work with ocf-data-sampler. This is Phase 2 of #103, building on the data fetching from PR #127.

The script transforms EIA data into the required format (time_utc, location_id dimensions) and estimates capacity from historical generation using the 99th percentile. Supports all major US regions and defaults to US48 to avoid duplicate data.

New files:

• preprocess_eia_data.py - preprocessing pipeline
• test_eia_preprocessing.py - tests (21 new tests)

Updated:

• training_model_new_country.md - added US usage examples

Fixes #103 (Phase 2)

How Has This Been Tested?

Added 21 tests covering schema transformation, capacity estimation, validation, and the full pipeline. All tests passing (31/31).

Ran manually with:
python -m open_data_pvnet.scripts.preprocess_eia_data
--start-date 2024-01-01
--end-date 2024-01-07
--regions CAISO
--output ./test.zarr

• Yes

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

• Yes - checked output schema and capacity values look reasonable

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

[mahendra-918]

@peterdudfield @siddharth7113 Hi! This PR implements Phase 2 of #103 - preprocessing EIA data for US solar generation.

I've addressed all the feedback from PR #127:

• Defaults to hourly data
• Uses US48 region to avoid duplicates
• Outputs to Zarr format
• Leverages the automatic pagination from the previous PR

Added 21 comprehensive tests (all passing). Would appreciate your review when you have a chance!

Let me know if anything needs changes. Thanks!