[BUG]: goroutine leak when leaderelection.NewLeaderElector return error

[alpha-baby]

https://github.com/kubernetes-sigs/controller-runtime/blob/main/pkg/manager/internal.go#L602

leak here : https://github.com/kubernetes-sigs/controller-runtime/blob/main/pkg/manager/internal.go#L519

When I was using controller-runtime v0.14.6 and configured an incorrect RenewDeadline setting, it caused an error with leaderelection.NewLeaderElector. Consequently, the following goroutine was continuously blocked:

1 @ 0x100a4f5a8 0x100a1793c 0x100a17664 0x1024ffeec 0x1024ffb78 0x1024ff320 0x1024ff12c 0x1025b51c0 0x100a80874
#	0x1024ffeeb	sigs.k8s.io/controller-runtime/pkg/manager.(*controllerManager).engageStopProcedure.func2+0x5b 		/Users/fujianhao/go/pkg/mod/sigs.k8s.io/controller-runtime@v0.14.6/pkg/manager/internal.go:575
#	0x1024ffb77	sigs.k8s.io/controller-runtime/pkg/manager.(*controllerManager).engageStopProcedure+0x657 	/Users/fujianhao/go/pkg/mod/sigs.k8s.io/controller-runtime@v0.14.6/pkg/manager/internal.go:615
#	0x1024ff31f	sigs.k8s.io/controller-runtime/pkg/manager.(*controllerManager).Start.func2+0x3f 		/Users/fujianhao/go/pkg/mod/sigs.k8s.io/controller-runtime@v0.14.6/pkg/manager/internal.go:432
#	0x1024ff12b	sigs.k8s.io/controller-runtime/pkg/manager.(*controllerManager).Start+0x77b 			/Users/fujianhao/go/pkg/mod/sigs.k8s.io/controller-runtime@v0.14.6/pkg/manager/internal.go:515
#	0x1025b51bf	main.main.func4+0x9f

Is this a bug? If so, I am interested in fixing it.