add fast mode toggle (#13212)

- add a local Fast mode setting in codex-core (similar to how model id
is currently stored on disk locally)
- send `service_tier=priority` on requests when Fast is enabled
- add `/fast` in the TUI and persist it locally
- feature flag

codex-rs/codex-api/src/common.rs

@@ -155,6 +155,8 @@ pub struct ResponsesApiRequest {
     pub stream: bool,
     pub include: Vec<String>,
     #[serde(skip_serializing_if = "Option::is_none")]
+    pub service_tier: Option<String>,
+    #[serde(skip_serializing_if = "Option::is_none")]
     pub prompt_cache_key: Option<String>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub text: Option<TextControls>,
@@ -174,6 +176,7 @@ impl From<&ResponsesApiRequest> for ResponseCreateWsRequest {
             store: request.store,
             stream: request.stream,
             include: request.include.clone(),
+            service_tier: request.service_tier.clone(),
             prompt_cache_key: request.prompt_cache_key.clone(),
             text: request.text.clone(),
             generate: None,
@@ -197,6 +200,8 @@ pub struct ResponseCreateWsRequest {
     pub stream: bool,
     pub include: Vec<String>,
     #[serde(skip_serializing_if = "Option::is_none")]
+    pub service_tier: Option<String>,
+    #[serde(skip_serializing_if = "Option::is_none")]
     pub prompt_cache_key: Option<String>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub text: Option<TextControls>,

codex-rs/codex-api/tests/clients.rs

@@ -265,6 +265,7 @@ async fn streaming_client_retries_on_transport_error() -> Result<()> {
         store: false,
         stream: true,
         include: Vec::new(),
+        service_tier: None,
         prompt_cache_key: None,
         text: None,
     };
@@ -306,6 +307,7 @@ async fn azure_default_store_attaches_ids_and_headers() -> Result<()> {
         store: true,
         stream: true,
         include: Vec::new(),
+        service_tier: None,
         prompt_cache_key: None,
         text: None,
     };