Add experimental thread prediction protocol types (#49480)

## What changed

Define `thread/prediction/request` with `threadId` and `sourceTurnId`, and `thread/prediction/updated` with a `completed` result containing optional text or a `failed` result.

Update generated JSON schemas, TypeScript types, and Python notification models. Route prediction notifications to their thread in the TUI and ignore them in the chat widget.

The request remains unimplemented and returns a method-not-found error.

GitOrigin-RevId: 09922f583d5c56f808d018121c651dcd92734515

codex-rs/app-server/src/message_processor.rs

@@ -11,6 +11,7 @@ use crate::current_time::app_server_time_provider;
 use crate::error_code::internal_error;
 use crate::error_code::invalid_params;
 use crate::error_code::invalid_request;
+use crate::error_code::method_not_found;
 use crate::extensions::ThreadExtensionDependencies;
 use crate::extensions::app_server_extension_event_sink;
 use crate::extensions::thread_extensions;
@@ -1443,6 +1444,9 @@ impl MessageProcessor {
                     .thread_unarchive(request_id.clone(), params)
                     .await
             }
+            ClientRequest::ThreadPredictionRequest { .. } => Err(method_not_found(
+                "thread/prediction/request is not implemented yet",
+            )),
             ClientRequest::ThreadCompactStart { params, .. } => {
                 self.thread_processor
                     .thread_compact_start(&request_id, params)