feat(rag): parallelize MinIO payload fetches for citations and VLM

[smasurekar]

Parallelize MinIO payload fetches for citations and VLM

Parallelizes MinIO get_payload calls used when building citations and VLM citation images. Adds a shared utility fetch_minio_payloads_for_documents() in utils/common.py that fetches all required payloads in parallel (capped at 10 workers via MINIO_FETCH_MAX_WORKERS). Response generator (prepare_citations) and VLM (citation image assembly) now use this utility instead of sequential per-doc MinIO calls, reducing latency when many image/structured docs are present.

Checklist

• I am familiar with the Contributing Guidelines.
• All commits are signed-off (git commit -s) and GPG signed (git commit -S).
• New or existing tests cover these changes.
• The documentation is up to date with these changes.
• If adjusting docker-compose.yaml environment variables have you ensured those are mimicked in the Helm values.yaml file.