Add audio output support to dynamic tools and code mode (#34080)

## What changed

- Add `inputAudio` content items to dynamic tool responses, app-server events, thread history, and generated protocol schemas.
- Add an `audio()` code-mode helper that accepts inline data URLs, audio URL objects, and MCP audio blocks.
- Convert MCP audio blocks into model input when audio is supported, and replace unsupported audio with an explanatory text item.
- Reject non-data audio URLs and track audio item counts in dynamic tool analytics.

## Testing

- Cover audio serialization, protocol round trips, thread-history conversion, MCP modality filtering, code-mode helper inputs, and invalid URL handling.

GitOrigin-RevId: 1ed52a8f9c62d4840fb71c5ec736b4a3566243d6

codex-rs/code-mode/src/runtime/callbacks.rs

@@ -5,6 +5,7 @@ use super::RuntimeEvent;
 use super::RuntimeState;
 use super::timers;
 use super::value::json_to_v8;
+use super::value::normalize_output_audio;
 use super::value::normalize_output_image;
 use super::value::serialize_output_text;
 use super::value::throw_type_error;
@@ -96,6 +97,26 @@ pub(super) fn text_callback(
     retval.set(v8::undefined(scope).into());
 }
 
+pub(super) fn audio_callback(
+    scope: &mut v8::PinScope<'_, '_>,
+    args: v8::FunctionCallbackArguments,
+    mut retval: v8::ReturnValue<v8::Value>,
+) {
+    let value = if args.length() == 0 {
+        v8::undefined(scope).into()
+    } else {
+        args.get(0)
+    };
+    let audio_item = match normalize_output_audio(scope, value) {
+        Ok(audio_item) => audio_item,
+        Err(()) => return,
+    };
+    if let Some(state) = scope.get_slot::<RuntimeState>() {
+        let _ = state.event_tx.send(RuntimeEvent::ContentItem(audio_item));
+    }
+    retval.set(v8::undefined(scope).into());
+}
+
 pub(super) fn image_callback(
     scope: &mut v8::PinScope<'_, '_>,
     args: v8::FunctionCallbackArguments,