feat(observability): enhance logging setup to route Python warnings and improve telemetry

feat: require live infrastructure for event store and inference, removing offline mode

feat: add Hugging Face inference backend next to Modal

feat: add optional mem0 cloud backend for memory index and update documentation

feat: unify model catalogue and self-hosted routing

feat: LiteLLM gateway routing profiles to Modal-served small models

feat: env-gated Postgres event store via SQLAlchemy (layered)

Refactor code structure for improved readability and maintainability

feat: phase 1 — memory layer, governor, second scenario, 70 passing tests