[FEATURE] Stream all steps of the backend process #5
Description
💡 Is your feature request related to a problem?
Yes, currently streaming is only done for significant steps in the memory and agents pipeline. This can result in bad user experience
✨ Describe the Solution
Every step of the backend process should be streamed to the frontend and displayed to the user. Essentially we have to yield the results from the FastAPI backend to the Electron application and display the changing states with NextJS
🔄 Alternatives Considered
We are thinking of incorporating reasoning like user experience in Sentient. But need a reasoning model for that which Llama3.2:3b doesn't support. So for now, streaming every step is the way to go
📌 Additional Context
No breaking changes needed