Enable server-side output truncation for Code Mode sessions (#57870)

## Why

Code Mode sessions always used client-side output handling, leaving server truncation settings and observation budgets unused.

## What changed

- Honor each session's `output_truncation` setting in the runtime and both host transports, and advertise the bounded-output capability.
- Apply `max_output_tokens` independently to execute, wait, and terminate responses in server mode. Keep client mode as the default when the setting is omitted.
- Use bounded output collection and rendering in server mode, preserving head and tail output with omission notices. Include script errors in the output budget while retaining failure status and a separate diagnostic capped at 64 KiB.
- Estimate audio tokens without the Tokio-dependent cache when rendering bounded output.

## Testing

Add runtime and host regression tests for session mode selection, per-request budgets, output overflow, oversized images, zero-token audio output, diagnostic preservation, and termination under output backpressure.

GitOrigin-RevId: 3d9661edcdd2dde31526bb92eb83fa0614f0fa88

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

@@ -144,6 +144,10 @@ pub enum RuntimeResponse {
     Result {
         cell_id: CellId,
         content_items: Vec<FunctionCallOutputContentItem>,
+        /// None means success; Some (including an empty string) means failure.
+        /// Client mode supplies the original diagnostic to append before truncation.
+        /// Server mode includes the diagnostic in the already-budgeted content and
+        /// supplies an independently bounded copy here for logging; do not append it again.
         error_text: Option<String>,
         /// Time between completion and retrieval when the result was buffered.
         #[serde(default, skip_serializing_if = "Duration::is_zero")]