[Bug] NPE throwed when recovering ledger: Cannot read field "distributionSchedule" because "this.lh" is null

### Search before asking

- [X] I searched in the [issues](https://github.com/apache/pulsar/issues) and found nothing similar.

### Version

Pulsar 2.11

### Minimal reproduce step

I don't know how to reproduce it

### What did you expect to see?

Recover ledger successfully

### What did you see instead?

9333 Oct 23 16:55:56 test_host pulsar[123890]: 2022-10-23T16:55:56,889+0800 [BookKeeperClientWorker-OrderedExecutor-1-0] WARN org.apache.pulsar.broker.service.BrokerSe
9333 rvice - Failed to create topic persistent://benchmark/ns-test/test-test-reusing-topic500-0000-partition-426
9334 Oct 23 16:55:56 test_host pulsar[123890]: org.apache.bookkeeper.mledger.ManagedLedgerException: Error while recovering ledger
9335 Oct 23 16:55:56 test_host pulsar[123890]: 2022-10-23T16:55:56,889+0800 [main-EventThread] INFO org.apache.bookkeeper.client.LedgerCreateOp - Ensemble: [10.225.50.
9335 43:3181, 10.224.145.38:3181, 10.225.105.4:3181] for ledger: 24474
9336 Oct 23 16:55:56 test_host pulsar[123890]: 2022-10-23T16:55:56,889+0800 [BookKeeperClientWorker-OrderedExecutor-26-0] ERROR org.apache.bookkeeper.common.util.SafeRu
9336 nnable - Unexpected throwable caught
9337 Oct 23 16:55:56 test_host pulsar[123890]: java.lang.NullPointerException: Cannot read field "distributionSchedule" because "this.lh" is null
9338 Oct 23 16:55:56 test_host pulsar[123890]: at org.apache.bookkeeper.client.PendingAddOp.safeRun(PendingAddOp.java:272) ~[org.apache.bookkeeper-bookkeeper-server-4.1 9338 6.1-SNAPSHOT.jar:4.16.1-SNAPSHOT]
9339 Oct 23 16:55:56 test_host pulsar[123890]: at org.apache.bookkeeper.common.util.SafeRunnable.run(SafeRunnable.java:36) ~[org.apache.bookkeeper-bookkeeper-common-4.1 9339 6.1-SNAPSHOT.jar:4.16.1-SNAPSHOT]
9340 Oct 23 16:55:56 test_host pulsar[123890]: at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136) ~[?:?]
9341 Oct 23 16:55:56 test_host pulsar[123890]: at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635) ~[?:?]
9342 Oct 23 16:55:56 test_host pulsar[123890]: at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30) ~[io.netty-netty-common-4.1.77.F 9342 inal.jar:4.1.77.Final]
9343 Oct 23 16:55:56 test_host pulsar[123890]: at java.lang.Thread.run(Thread.java:833) ~[?:?]

### Anything else?

_No response_

### Are you willing to submit a PR?

- [X] I'm willing to submit a PR!

Start with org.apache.bookkeeper.client.PendingAddOp.safeRun at PendingAddOp.java:272 and trace the ledger-recovery path that reaches it with a null lh. Use the reported Pulsar 2.11 and BookKeeper 4.16.1-SNAPSHOT context to investigate the failure, since the issue provides no reproduction steps. Done means ledger recovery completes without this NullPointerException.

Written by the indexing model from the issue text.