The recent paper “Solving A Million-Step LLM Task with Zero Errors” (https://arxiv.org/pdf/2511.09030) proposes a new and simple agentic architecture to improve performance on long running tasks. Are there any plans to integrate MAKER or MDAP abstractions into LangChain/LangGraph? I can envision specialized MAKER nodes being beneficial, especially if combined with methods to automatically “massively decompose” tasks for those MAKER nodes