Fix the seek method could be blocked forever when subscribe RPC is slower than seek RPC

[BewareMyPower]

Fixes #534

Use mutex_ to synchronize the following two composited tasks:

1. Call setCnx, cnx->registerConsumer and clearReceiveQueue directly block in connectionOpened. Once it's detected the status is COMPLETED, delay the callback completion after the subscribe response is processed.
2. When receiving a seek response, only set the status to COMPLETED when the reconnection is not done.

Since now many accesses to message id fields are also protected by the mutex_, remove mutexForMessageId_ and protect other accesses by mutex_ as well.

Add more debug logs to ClientConnection and support mocking some requests with specific latencies, as well as new tests testSubscribeSeekRaces and testReconnectionSlow to cover all possible cases.

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[Copilot]

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated 8 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[BewareMyPower]

cc @gaoran10 as well since you also have the context