DB Corruption Panic on Sled 0.34 After Stopping Container

A Sled database (version 0.34) enters a corrupt state and cannot be opened after the Docker container using it was abruptly stopped. When attempting to restart the application, it panics with a Corruption error.

---

Current Behavior

When the application starts, the following output is observed in the logs before the process terminates:

1. A WARN message indicating a corrupt or empty snapshot file:
   复制

   WARN sled::pagecache::snapshot: empty/corrupt snapshot file found

2. The application panics when calling .unwrap() on a corruption error result:
   复制

   thread 'main' panicked at ... called `Result::unwrap()` on an `Err` value: Corruption { at: None, bt: () }

---

Expected Behavior

Sled should ideally be able to recover from an unexpected shutdown. Barring that, it should provide a clear recovery path or tooling to repair the database and salvage data.