Integrate BacDive API fetching into kg download workflow

[turbomam]

Current State

BacDive data is currently fetched from a Google Drive file (see PR #349). The download_bacdive.py utility script exists (PR #273, thanks @realmarcin) and was updated to use .env credentials (PR #314), but it's not integrated into the standard kg download workflow.

Proposal

Integrate BacDive API fetching into kg download, similar to how MediaDive bulk download works (see _post_download_mediadive_bulk() in download.py).

This would make the build fully reproducible from source for users with BacDive API credentials.

Notes

• The current download_bacdive.py scans 200k IDs sequentially (~97k actually exist) - may need optimization
• Please correct me if there's existing integration I'm overlooking
• Related: identify python libraries that are necessary for core tasks like kg download but are not specified in pyproject.toml #333, Is DSMZ "giving us" the bacdive_strains.json file, or are we scraping it with https://pypi.org/project/bacdive/ ? #335

Acceptance Criteria

• kg download can fetch BacDive data via API (with credentials)
• Documentation for credential setup
• Fallback behavior if credentials unavailable