Enable uploaded audio with durable rejection recovery (#59234)

## Why

Uploaded audio was rejected even for audio-capable models. Invalid or expired recordings also need to be excluded from later requests so users can continue the conversation.

## What changed

- Accept uploaded audio when the selected model supports audio; otherwise, preserve valid sibling input and report the unsupported attachment before inference.
- Track successful audio decoding by model and provider, and attribute provider rejections against the exact submitted request. Preserve attachments when attribution is ambiguous.
- Persist confirmed rejected file IDs across compaction and resume, omitting them from future requests without rewriting retained user history.
- Check audio availability before remote compaction and retry once without an unambiguously rejected recording. Include guidance to remove or replace affected attachments in audio errors.

## Testing

Add unit and scenario coverage for uploaded audio across follow-up turns, compaction, cold and paginated resume, model switching, ambiguous rejections, expired files, and queued or interrupted input.

GitOrigin-RevId: e701901e7b1c0865c9f2a5918197215f549b06c0

codex-rs/codex-api/src/api_bridge.rs

@@ -86,9 +86,13 @@ fn map_api_error_details(err: ApiError) -> CodexErr {
             })
         }
         ApiError::InvalidRequest { message } => CodexErr::InvalidRequest(message),
-        ApiError::InvalidAudioInput { message } => {
-            CodexErr::new(CodexErrorDetails::InvalidAudioInput { message })
-        }
+        ApiError::InvalidAudioInput {
+            message,
+            per_recording,
+        } => CodexErr::new(CodexErrorDetails::InvalidAudioInput {
+            message,
+            per_recording,
+        }),
         ApiError::ModelNotFound { message, status } => {
             CodexErr::new(CodexErrorDetails::ModelNotFound { message, status })
         }

codex-rs/codex-api/src/error.rs

@@ -42,7 +42,11 @@ pub enum ApiError {
     #[error("invalid request: {message}")]
     InvalidRequest { message: String },
     #[error("{message}")]
-    InvalidAudioInput { message: String },
+    InvalidAudioInput {
+        message: String,
+        /// True when the provider rejected an individual clip, independent of request size.
+        per_recording: bool,
+    },
     #[error("{message}")]
     ModelNotFound {
         message: String,
@@ -93,6 +97,7 @@ pub(crate) fn parse_terminal_provider_error(
             | "missing_audio_sample_rate",
             None | Some(StatusCode::BAD_REQUEST),
         ) => Some(ApiError::InvalidAudioInput {
+            per_recording: code != "audio_too_large",
             message: error
                 .get("message")
                 .and_then(Value::as_str)