router.NewConsistentHashPool has error occurred while closing

[remainbeen]

Describe the bug
I used router.NewConsistentHashPool to create a pool in cluster. Then when I closed the program, I got an error:
`7:46PM INF [ACTOR] Recovering lib=Proto.Actor system=fBr5nhw5vuR79aeLVBJe3V self="Address:"127.0.0.1:64236" Id:"partition-activator"" reason="runtime error: invalid memory address or nil pointer dereference"
debug.Stack(): goroutine 27 [running]:
runtime/debug.Stack()
C:/Program Files/Go/src/runtime/debug/stack.go:26 +0x5e
github.com/asynkron/protoactor-go/actor.(*actorContext).EscalateFailure(0xc000200900, {0x11ada40, 0x18d9d60}, {0x1236760, 0xc0006b2500})
D:/Study/protoactor-go/actor/actor_context.go:709 +0x19b
github.com/asynkron/protoactor-go/actor.(*defaultMailbox).run.func1()
D:/Study/protoactor-go/actor/mailbox.go:132 +0x58
panic({0x11ada40?, 0x18d9d60?})
C:/Program Files/Go/src/runtime/panic.go:785 +0x132
github.com/asynkron/protoactor-go/cluster/identitylookup/disthash.(*placementActor).onTerminated(0xc0002082d0, 0xc0006b2500)
D:/Study/protoactor-go/cluster/identitylookup/disthash/placement_actor.go:51 +0xe0
github.com/asynkron/protoactor-go/cluster/identitylookup/disthash.(*placementActor).Receive(0xc0002082d0, {0x13c86a8, 0xc000200900})
D:/Study/protoactor-go/cluster/identitylookup/disthash/placement_actor.go:39 +0x1fe
github.com/asynkron/protoactor-go/actor.(*actorContext).defaultReceive(0xc000200900)
D:/Study/protoactor-go/actor/actor_context.go:372 +0x98
github.com/asynkron/protoactor-go/actor.(*actorContext).processMessage(0xc000200900, {0x1236760, 0xc0006b2500})
D:/Study/protoactor-go/actor/actor_context.go:527 +0xf9
github.com/asynkron/protoactor-go/actor.(*actorContext).InvokeUserMessage(0xc000200900, {0x1236760, 0xc0006b2500})
D:/Study/protoactor-go/actor/actor_context.go:505 +0x465
github.com/asynkron/protoactor-go/actor.(*actorContext).handleTerminated(0xc000200900, 0xc0006b2500?)
D:/Study/protoactor-go/actor/actor_context.go:628 +0x49
github.com/asynkron/protoactor-go/actor.(*actorContext).InvokeSystemMessage(0xc000200900?, {0x1236760?, 0xc0006b2500?})
D:/Study/protoactor-go/actor/actor_context.go:561 +0x15e
github.com/asynkron/protoactor-go/actor.(*defaultMailbox).run(0xc000200960)
D:/Study/protoactor-go/actor/mailbox.go:154 +0x1a6
github.com/asynkron/protoactor-go/actor.(*defaultMailbox).processMessages(0xc000200960)
D:/Study/protoactor-go/actor/mailbox.go:105 +0x25
created by github.com/asynkron/protoactor-go/actor.goroutineDispatcher.Schedule in goroutine 1
D:/Study/protoactor-go/actor/dispatcher.go:13 +0x16

[Supervision] Actor: Address:"127.0.0.1:64236" Id:"partition-activator" failed with message: who:{Address:"127.0.0.1:64236" Id:"partition-activator/test$d/router"} exception: runtime error: invalid memory address or nil pointer dereference`

To Reproduce
The test code is in the attachment。
1 build and run it.
2 press CTRL+C to close it.
You will see the error ocurred。

main.go.txt

[remainbeen]

I found the error occurred here:

`func (p *placementActor) onTerminated(msg *actor.Terminated) {
found, key, meta := p.pidToMeta(msg.Who)
clusterKind := p.cluster.GetClusterKind(meta.ID.Kind)
clusterKind.Dec()

activationTerminated := &clustering.ActivationTerminated{
	Pid:             msg.Who,
	ClusterIdentity: meta.ID,
}
p.partitionManager.cluster.MemberList.BroadcastEvent(activationTerminated, true)

if found {
	delete(p.actors, *key)
}

}`

The reason for the error is that the pool's pid is not registered in p.actors, so pidToMeta returns false and nil. No judgment is made here, which is incorrect behavior.