At this point, the pod and all containers have fully shut down, and we can tear everything down.

Because the application(s) are already entirely gone, there is no need to have any "draining" time -- we can immediately terminate.
Furthermore, because the Pod is torn down, we cannot even send any data at all (at least on most clusters, the `veth` is torn down upon Pod deletion).

         *
         * 2a. If the resolved types are identical to the original types, then we can return the
         * original, identical JDK TypeVariable. By doing so, we sidestep the problem entirely.
         *
         * 2b. If the resolved types are different from the original types, things are trickier. The
         * only way to get a TypeVariable instance for the resolved types is to create our own. The

            // public interface or class (because the actual implementing
            // class may be a facade...)

            Method publicMethod = getPublicMethod(method);

            // it is entirely possible that there is no public method for
            // the methods of this class (i.e. in the facade, a method
            // that isn't on any of the interfaces or superclass

  // +optional
  optional string reason = 2;

  // EvaluationError is an indication that some error occurred during the authorization check.
  // It is entirely possible to get an error and be able to continue determine authorization status in spite of it.
  // For instance, RBAC can be missing a role, but enough roles are still present and bound to reason about the request.
  // +optional

	// because we are using `List()` in the handler, without this requeue,
	// the fake client will sometimes drop pod events leading to test flakes.
	//
	// WaitForCacheSync *helps*, but does not entirely fix this problem
	s.namespaces = kclient.New[*corev1.Namespace](kubeClient)
	s.namespaces.AddEventHandler(controllers.FromEventHandler(func(o controllers.Event) {
		s.queue.Add(o)
	}))

	return s
}

  // +optional
  optional string reason = 2;

  // EvaluationError is an indication that some error occurred during the authorization check.
  // It is entirely possible to get an error and be able to continue determine authorization status in spite of it.
  // For instance, RBAC can be missing a role, but enough roles are still present and bound to reason about the request.
  // +optional

   *
   * @since 8.0
   */
  public static final byte SOH = 1;

  /**
   * Start of Text: A communication control character which precedes a sequence of characters that
   * is to be treated as an entity and entirely transmitted through to the ultimate destination.
   * Such a sequence is referred to as "text." STX may be used to terminate a sequence of characters
   * started by SOH.
   *
   * @since 8.0
   */

So let's focus on the main cases, where it could crash entirely in some particular cases **in the future**, and it still makes sense to restart it.

///

         * ReflectionFreeAssertThrowsTest with a testNulls method, but that's annoying to have to do
         * for a package-private utility class. So we skip the class entirely instead.
         */
        continue;
      }
      try {
        tester.doTestNulls(classToTest, visibility);
      } catch (Throwable e) {

         *
         * 2a. If the resolved types are identical to the original types, then we can return the
         * original, identical JDK TypeVariable. By doing so, we sidestep the problem entirely.
         *
         * 2b. If the resolved types are different from the original types, things are trickier. The
         * only way to get a TypeVariable instance for the resolved types is to create our own. The