fix: record original upstream status code when failover exhausted (#1128)

When all failover accounts are exhausted, handleFailoverExhausted maps
the upstream status code (e.g. 403) to a client-facing code (e.g. 502)
but did not write the original code to the gin context. This caused ops
error logs to show the mapped code instead of the real upstream code.

Call SetOpsUpstreamError before mapUpstreamError in all failover-
exhausted paths so that ops_error_logger captures the true upstream
status code and message.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

backend/internal/handler/sora_gateway_handler.go

@@ -484,6 +484,9 @@ func (h *SoraGatewayHandler) handleConcurrencyError(c *gin.Context, err error, s
 }
 
 func (h *SoraGatewayHandler) handleFailoverExhausted(c *gin.Context, statusCode int, responseHeaders http.Header, responseBody []byte, streamStarted bool) {
+	upstreamMsg := service.ExtractUpstreamErrorMessage(responseBody)
+	service.SetOpsUpstreamError(c, statusCode, upstreamMsg, "")
+
 	status, errType, errMsg := h.mapUpstreamError(statusCode, responseHeaders, responseBody)
 	h.handleStreamingAwareError(c, status, errType, errMsg, streamStarted)
 }