feat(openai): implement Responses API streaming and proactive routing

This commit adds support for the OpenAI Responses API in both streaming and non-streaming modes. It also implements proactive routing for gpt-5 and codex models and cleans up unused 'session' variable warnings across the dashboard source files.

src/server/mod.rs

@@ -226,7 +226,15 @@ async fn chat_completions(
         let prompt_tokens = crate::utils::tokens::estimate_request_tokens(&model, &unified_request);
 
         // Handle streaming response
-        let stream_result = provider.chat_completion_stream(unified_request).await;
+        // Allow provider-specific routing for streaming too
+        let use_responses = provider.name() == "openai"
+            && crate::utils::registry::model_prefers_responses(&state.model_registry, &unified_request.model);
+
+        let stream_result = if use_responses {
+            provider.chat_responses_stream(unified_request).await
+        } else {
+            provider.chat_completion_stream(unified_request).await
+        };
 
         match stream_result {
             Ok(stream) => {