AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play | Read Paper on Bytez