[server] local-cluster.sh starts successfully but processes fail silently when JAVA_HOME is misconfigured

### Search before asking

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

### Fluss version

0.9.0 (latest release)

### Please describe the bug 🐞

When running `./bin/local-cluster.sh start` with an incorrect or missing `JAVA_HOME` configuration, the
script reports "Starting cluster." and exits with success, but the actual JVM processes fail immediately
with a `UnsupportedClassVersionError`. This gives users a false impression that the cluster started
successfully.

### Steps to Reproduce

1. Ensure `JAVA_HOME` is either unset or points to a Java 8 runtime (while Fluss requires Java 11+)
2. Run `./bin/local-cluster.sh start`
3. Observe the output showing successful startup messages
4. Check `log/fluss-*-coordinator-server-*.out` to find the actual JVM crash

### Solution

_No response_

### Are you willing to submit a PR?

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