Improve data retrieval and workflow robustness

[danielelerede-oet]

Changes proposed in this Pull Request

This PR consolidates fixes identified while running the electricity and sector-coupled workflows.

• Download the required PUDL Parquet files through explicit Snakemake rules and store them in a versioned local cache, avoiding direct remote access from DuckDB.
• Pass the local PUDL inputs consistently to the power plant, demand, stock, and cost-building scripts.
• Improve Zenodo downloads by supporting archived renewable-profile files, creating missing parent directories, validating downloads, and cleaning up incomplete or corrupted files.
• Update obsolete NLR Data Catalog URLs.
• Handle capacity plotting when a single- or multi-horizon network has no new extendable capacity.
• Improve ERA5 cutout download reliability by using monthly, sequential requests.
• Fix natural-gas data handling by selecting the requested year before temporal remapping and aggregating duplicate trade entries.
• Accept both U.S. and U.S. Total in EIA gas-storage data.

The branch has been rebased onto develop; run-specific 2023 configuration files and downloaded PUDL data are not included.

Checklist

• I tested my contribution locally and it seems to work fine.
• Code and workflow changes are sufficiently documented.
• Changed dependencies are added to envs/environment.yaml.
• Changes in configuration options are added in all of config.default.yaml.
• Changes in configuration options are also documented in doc/configtables/*.csv.