🐛 BUG: LLM Chat template does not implement SSE streaming as claimed

### Which template does this pertain to?

llm-chat-app-template

### What versions are you using?

Chrome Version 140.0.7339.214

### What operating system and version are you using?

Mac Sonoma 14.7.1

### Please provide a link to a minimal reproduction (optional)

_No response_

### Describe the Bug

The template claims to stream the AI model’s response using SSE, but the source code doesn’t use the EventSource class. Even on the demo page, the text appears all at once—regardless of length—instead of streaming in real time, piece by piece.

### Please provide any relevant error logs

_No response_