dra e2e: fix stack unwinding in helper function

When failing inside the `ginkgo.By` callback function, skipping intermediate
stack frames didn't work properly because `ginkgo.By` itself and other internal
code is also on the stack.

To fix this, the code which can fail now runs outside of such a
callback. That's not a big loss, the only advantage of the callback was getting
timing statistics from Ginkgo which weren't used in practice.

test/e2e/dra/dra.go

@@ -1128,7 +1128,7 @@ func (b *builder) podExternalMultiple() *v1.Pod {
 func (b *builder) create(ctx context.Context, objs ...klog.KMetadata) []klog.KMetadata {
 	var createdObjs []klog.KMetadata
 	for _, obj := range objs {
-		ginkgo.By(fmt.Sprintf("creating %T %s", obj, obj.GetName()), func() {
+		ginkgo.By(fmt.Sprintf("creating %T %s", obj, obj.GetName()))
 		var err error
 		var createdObj klog.KMetadata
 		switch obj := obj.(type) {
@@ -1147,7 +1147,6 @@ func (b *builder) create(ctx context.Context, objs ...klog.KMetadata) []klog.KMe
 		}
 		framework.ExpectNoErrorWithOffset(1, err, "create %T", obj)
 		createdObjs = append(createdObjs, createdObj)
-		})
 	}
 	return createdObjs
 }