The interactive configuration wizard (--init) lists available LLM providers but does not include the orchestrator option.
Since orchestrator is always-on and allows multi-model routing, it should be presented as a provider choice during initial setup.
Current behavior: Provider selection offers Ollama, Claude, OpenAI, Candle (if feature-gated).
Expected behavior: Orchestrator should appear in the provider list, allowing users to configure multi-model routing from the start.