func (s *meshDataplane) AddPodToMesh(ctx context.Context, pod *corev1.Pod, podIPs []netip.Addr, netNs string) error {
	var retErr error
	err := s.netServer.AddPodToMesh(ctx, pod, podIPs, netNs)
	if err != nil {
		log.Errorf("failed to add pod to ztunnel: %v", err)
		if !errors.Is(err, ErrPartialAdd) {
			return err
		}
		retErr = err
	}

	log.Debugf("annotating pod %s", pod.Name)

		"--mark", inpodMark,
		"-j", "CONNMARK",
		"--set-xmark", inpodTproxyMark)

	// Handle healthcheck probes from the host node. In the host netns, before the packet enters the pod, we SNAT
	// the healthcheck packet to a fixed IP if the packet is coming from a node-local process with a socket.
	//

        assumes layers only ever have one positional argument. This assumption
        doesn't hold true before TF 2.4 either, but is more likely to cause
        issues now.
    *   Code that manually enters `keras.backend.get_graph()` before building a
        functional model is no longer needed.
    *   Start enforcing input shape assumptions when calling Functional API

this release is **compatibility** with Pydantic v1 and v2, to make sure your current apps keep working. Later there will be more focus on refactors, correctness, code improvements, and then **performance** improvements. Some third-party early beta testers that ran benchmarks on the beta releases of FastAPI reported improvements of **2x - 3x**. Which is not bad for just doing `pip install --upgrade fastapi pydantic`. This was not an official benchmark and I didn't check it myself, but it's a good sign....