[BUG]: High CPU usage when opening very long conversation

### How are you running AnythingLLM? Windows Desktop App ### What happened? Opening one specific very long conversation causes CPU usage to spike to around 96%. Other conversations in the same workspace remain around 5–10% CPU usage. Observations: Workspace contains approximately 530 vectors. CPU usage appears to correlate with conversation length rather than vector count. Switching away from the long conversation immediately reduces CPU usage. The issue occurs consistently and is reproducible. Expected behavior: Opening a long conversation should not cause excessive CPU usage. Actual behavior: CPU usage spikes to approximately 96% when the long conversation is opened. Environment: AnythingLLM version: [v1.13.0] Browser: [Chrome 148.0.7778.217] OS: [Windows 11] CPU: [AMD Ryzen AI 9 HX 370 w/ Radeon 890M] ### Are there known steps to reproduce? Steps to reproduce: 1. Open a workspace containing approximately 530 vectors. 2. Open a very long conversation with a large chat history. 3. Observe CPU usage in Task Manager. 4. Switch to a shorter conversation in the same workspace. 5. Compare CPU usage. Result: CPU usage rises to approximately 96% when the long conversation is open and drops back to around 5–10% when switching to a shorter conversation. Additional note: The issue appears to be related to conversation length rather than the number of vectors in the workspace. ### LLM Provider & Model (if applicable) Ollama / gemma4:31b ### Embedder Provider & Model (if applicable) Ollama / nomic-embed-text-1