pub async fn stream_response(
builder: RequestBuilder,
provider: &str,
call_id: &str,
sink: &dyn LlmStreamSink,
) -> Result<CloudAiResponse, CloudAiError>Expand description
Stream an OpenAI-compatible Chat Completions response, emitting each
content delta through sink. Returns the accumulated full response
once the stream closes so the caller still gets a CloudAiResponse
to surface in the node output.
Handles the canonical SSE shape (one data: <json>\n per chunk plus
a terminating data: [DONE]). Servers that don’t include a usage
chunk leave token counts at zero, mirroring the non-streaming path.