feat(storage): configure object storage and vector database

Summary

• Add a real S3-compatible object storage client with signed put, get, head, delete, public URL, and bucket helpers on top of the existing deployment/runtime configuration.
• Wire runtime object storage and opendata-vector endpoint config through HIVE_S3_* and HIVE_OPENDATA_VECTOR_URL.
• Add an optional Helm-managed opendata-vector deployment that reuses Hive’s object storage credentials and bucket, storing its index under vector.storage.prefix with a PVC cache.
• Enable Tuist production opendata-vector values while documenting the portable self-hosted setup.

Testing

• mix format
• mix test test/hive/object_storage_test.exs test/hive/opendata_vector_test.exs
• mix compile --warnings-as-errors
• helm lint infra/helm/hive
• helm template hive infra/helm/hive
• helm template hive infra/helm/hive -f infra/helm/hive/values-production.yaml
• helm template hive infra/helm/hive --set env.HIVE_OBJECT_STORAGE_PROVIDER=s3 --set env.HIVE_S3_BUCKET=hive-objects --set env.HIVE_S3_REGION=us-east-1 --set env.HIVE_S3_ENDPOINT_URL=https://s3.example.com --set vector.enabled=true --set vector.storage.prefix=vector
• mix test

No GitHub comments yet.