panic during node sighup processing in case of problems with rpc node

[evgeniiz321]

info	neofs-node/config.go:902	SIGHUP has been received, rereading configuration...
info	shard/control.go:315	trying to restore read-write mode	{"shard_id": "EgZdmowYzd4UqrK63zvAFg"}
info	shard/mode.go:28	setting shard mode	{"shard_id": "EgZdmowYzd4UqrK63zvAFg", "old_mode": "READ_WRITE", "new_mode": "READ_WRITE"}
info	shard/mode.go:69	shard mode set successfully	{"shard_id": "EgZdmowYzd4UqrK63zvAFg", "mode": "READ_WRITE"}
info	shard/control.go:315	trying to restore read-write mode	{"shard_id": "WQ1oGrsfXW2xe54Uy5jcKS"}
info	shard/mode.go:28	setting shard mode	{"shard_id": "WQ1oGrsfXW2xe54Uy5jcKS", "old_mode": "READ_WRITE", "new_mode": "READ_WRITE"}
info	shard/mode.go:69	shard mode set successfully	{"shard_id": "WQ1oGrsfXW2xe54Uy5jcKS", "mode": "READ_WRITE"}
info	client/notifications.go:228	RPC connection lost, attempting reconnect
info	neofs-node/config.go:960	configuration has been reloaded successfully
info	client/multi.go:53	connection to RPC node has been established	{"endpoint": "ws://localhost:58302/ws"}
info	neofs-node/main.go:207	restarting internal services because of RPC connection loss...
info	event/listener.go:221	stop event listener by context	{"reason": "context canceled"}
warn	neofs-node/main.go:181	internal application error	{"message": "restarting after morph connection was lost: getting network state: could not get current epoch number: could not perform test invocation (epoch): chain/client: contract execution finished with state FAULT; exception: at instruction 31 (SYSCALL): System.Contract.Call failed: called contract 4ea20d322984f445a4a737dbbd4bb1900cdee905 not found: key not found"}
info	neofs-node/grpc.go:138	stopping gRPC server...	{"name": "NeoFS Control API"}
error	client/notifications.go:267	could not restore block subscription	{"error": "connection lost: while waiting for the response"}
info	client/notifications.go:228	RPC connection lost, attempting reconnect
info	neofs-node/grpc.go:154	gRPC server stopped successfully	{"name": "NeoFS Control API"}
info	neofs-node/storage.go:65	closing components of the storage engine...
warn	client/multi.go:43	could not establish connection to RPC node	{"endpoint": "ws://localhost:58302/ws", "error": "WS client creation: dial tcp: lookup localhost: operation was canceled"}
debug	blobstor/control.go:44	closing...	{"shard_id": "EgZdmowYzd4UqrK63zvAFg"}
info	shard/gc.go:165	waiting for GC workers to stop...
debug	shard/gc.go:151	GC is stopped
warn	shard/gc.go:105	stop event listener by closed channel
debug	blobstor/control.go:44	closing...	{"shard_id": "WQ1oGrsfXW2xe54Uy5jcKS"}
info	shard/gc.go:165	waiting for GC workers to stop...
debug	shard/gc.go:151	GC is stopped
warn	shard/gc.go:105	stop event listener by closed channel
info	neofs-node/storage.go:73	all components of the storage engine closed successfully
info	neofs-node/grpc.go:138	stopping gRPC server...	{"name": "NeoFS Public API"}
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x20 pc=0x1009fed98]

goroutine 70 [running]:
github.com/nspcc-dev/neofs-node/pkg/morph/client.(*Client).routeNotifications(0x140001be240)
	github.com/nspcc-dev/neofs-node/pkg/morph/client/notifications.go:236 +0xf8
created by github.com/nspcc-dev/neofs-node/pkg/morph/client.New in goroutine 1
	github.com/nspcc-dev/neofs-node/pkg/morph/client/constructor.go:172 +0x630
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x0 pc=0x1009ff044]

goroutine 372 [running]:
github.com/nspcc-dev/neofs-node/pkg/morph/client.(*Client).restoreSubscriptions(0x140001be240, 0x0, 0x140001172c0)
	github.com/nspcc-dev/neofs-node/pkg/morph/client/notifications.go:265 +0xc4
created by github.com/nspcc-dev/neofs-node/pkg/morph/client.(*Client).routeNotifications in goroutine 70
	github.com/nspcc-dev/neofs-node/pkg/morph/client/notifications.go:230 +0xe4

See nspcc-dev/neofs-testcases#915