Preserve terminal turn errors in replayed history (#35524)

## Why

When rebuilding a thread from rollout events, errors embedded in turn completion
events were ignored. Failed retries could therefore be restored as completed
turns, causing warnings such as model-overload errors to disappear from the TUI
transcript.

## What changed

- Mark turns with terminal completion errors as failed and retain their message
  and structured error information.
- Apply late completion errors to their matching historical turn without
  interrupting the active turn.

## Testing

Added history-builder coverage for current and late completion errors, plus a
TUI replay snapshot covering repeated overload failures.

GitOrigin-RevId: 5c474baa0fd5c73368b444d79fb850cbd4c8023a

codex-rs/tui/src/chatwidget/tests/history_replay.rs

@@ -75,6 +75,54 @@ async fn resumed_initial_messages_render_history() {
     );
 }
 
+#[tokio::test]
+async fn replayed_failed_turns_preserve_overload_warnings_between_retries() {
+    let (mut chat, mut rx, _ops) = make_chatwidget_manual(/*model_override*/ None).await;
+    let prompt = "The workspace also looks super confusing with its separator.";
+    let error_message = "Selected model is at capacity. Please try a different model.";
+    let failed_turn = |turn_id: &str, item_id: &str| AppServerTurn {
+        items: vec![AppServerThreadItem::UserMessage {
+            id: item_id.to_string(),
+            client_id: None,
+            content: vec![AppServerUserInput::Text {
+                text: prompt.to_string(),
+                text_elements: Vec::new(),
+            }],
+        }],
+        ..app_server_turn(
+            turn_id,
+            AppServerTurnStatus::Failed,
+            /*duration_ms*/ None,
+            /*error*/
+            Some(AppServerTurnError {
+                message: error_message.to_string(),
+                codex_error_info: Some(CodexErrorInfo::ServerOverloaded),
+                additional_details: None,
+            }),
+        )
+    };
+
+    chat.replay_thread_turns(
+        vec![
+            failed_turn("turn-1", "user-1"),
+            failed_turn("turn-2", "user-2"),
+        ],
+        ReplayKind::ResumeInitialMessages,
+    );
+
+    let rendered = drain_insert_history(&mut rx)
+        .into_iter()
+        .map(|lines| lines_to_single_string(&lines))
+        .collect::<String>();
+
+    assert_eq!(rendered.matches(prompt).count(), 2);
+    assert_eq!(rendered.matches(error_message).count(), 2);
+    insta::assert_snapshot!(
+        "replayed_failed_turns_preserve_overload_warnings_between_retries",
+        rendered
+    );
+}
+
 #[tokio::test]
 async fn restored_conversation_ultra_remains_selected_after_switching_to_plan() {
     let (mut chat, _rx, _ops) = make_chatwidget_manual(Some("gpt-5.4")).await;

codex-rs/tui/src/chatwidget/tests/snapshots/codex_tui__chatwidget__tests__history_replay__replayed_failed_turns_preserve_overload_warnings_between_retries.snap

@@ -0,0 +1,15 @@
+---
+source: tui/src/chatwidget/tests/history_replay.rs
+expression: rendered
+---
+
+› The workspace also looks super confusing with its separator.
+
+
+⚠ Selected model is at capacity. Please try a different model.
+
+
+› The workspace also looks super confusing with its separator.
+
+
+⚠ Selected model is at capacity. Please try a different model.