Moved `InternalMessage` into schemas and created a `BeliefMessage`
model. Also added the ability for agents to communicate via ZMQ to
agents on another process.
ref: N25B-316
By efficiently checking when the next work has to be done, we can
increase performance not having to "busy loop". Time from transcription
-> message to LLM agent is now down to sub 1 millisecond.
ref: N25B-316
