fix: call inflight.mark_error() in media handler error paths (#7917)

Signed-off-by: Matej Kosec <mkosec@nvidia.com>

lib/llm/src/http/service/openai.rs

@@ -1998,7 +1998,9 @@ async fn images(
                 .metrics_clone()
                 .inc_rejection(&model, super::metrics::Endpoint::Images);
         }
-        ErrorMessage::from_anyhow(e, "Failed to generate images")
+        let err_response = ErrorMessage::from_anyhow(e, "Failed to generate images");
+        inflight.mark_error(extract_error_type_from_response(&err_response));
+        err_response
     })?;
 
     // Process stream to collect metrics and drop http_queue_guard on first response
@@ -2018,7 +2020,9 @@ async fn images(
         .await
         .map_err(|e| {
             tracing::error!("Failed to fold images stream for {}: {:?}", request_id, e);
-            ErrorMessage::internal_server_error("Failed to fold images stream")
+            let err_response = ErrorMessage::internal_server_error("Failed to fold images stream");
+            inflight.mark_error(extract_error_type_from_response(&err_response));
+            err_response
         })?;
 
     inflight.mark_ok();
@@ -2085,7 +2089,9 @@ async fn videos(
                 .metrics_clone()
                 .inc_rejection(&model, super::metrics::Endpoint::Videos);
         }
-        ErrorMessage::from_anyhow(e, "Failed to generate videos")
+        let err_response = ErrorMessage::from_anyhow(e, "Failed to generate videos");
+        inflight.mark_error(extract_error_type_from_response(&err_response));
+        err_response
     })?;
 
     // Process stream to collect metrics and drop http_queue_guard on first token
@@ -2105,7 +2111,9 @@ async fn videos(
         .await
         .map_err(|e| {
             tracing::error!("Failed to fold videos stream for {}: {:?}", request_id, e);
-            ErrorMessage::internal_server_error("Failed to fold videos stream")
+            let err_response = ErrorMessage::internal_server_error("Failed to fold videos stream");
+            inflight.mark_error(extract_error_type_from_response(&err_response));
+            err_response
         })?;
 
     inflight.mark_ok();
@@ -2150,7 +2158,9 @@ async fn video_stream(
                 .metrics_clone()
                 .inc_rejection(&model, super::metrics::Endpoint::Videos);
         }
-        ErrorMessage::from_anyhow(e, "Failed to start video stream")
+        let err_response = ErrorMessage::from_anyhow(e, "Failed to start video stream");
+        inflight.mark_error(extract_error_type_from_response(&err_response));
+        err_response
     })?;
 
     // Capture the context to cancel the stream if the client disconnects.
@@ -2234,6 +2244,7 @@ async fn video_stream(
                 }
                 _ = ctx.stopped() => {
                     tracing::trace!("Context stopped; breaking MJPEG stream");
+                    inflight.mark_error(ErrorType::Cancelled);
                     break;
                 }
             }
@@ -2249,6 +2260,8 @@ async fn video_stream(
         .body(Body::from_stream(monitored_stream))
         .map(|r| r.into_response())
         .map_err(|e| {
+            // inflight is already owned by the monitored_stream which handles
+            // mark_ok (stream end) and mark_error (cancellation).
             ErrorMessage::internal_server_error(&format!("Failed to build MJPEG response: {e}"))
         })
 }