RATIS-2220. Skip further tests after leak detected

[adoroszlai]

What changes were proposed in this pull request?

After the first test case has failed, the leak detector will fail the following test cases (even if there are no new leaks) since it won't reset the leak trackers. (comment)

We should skip further tests to avoid wasting time.

(I'm guessing JUnit4 tests may not benefit from this, since it's added in a JUnit5 hook method.)

https://issues.apache.org/jira/browse/RATIS-2220

How was this patch tested?

10x10 run of TestRaftWithGrpc, as expected has many failures.

In each case at most 1 test failed with error, various number of other tests were skipped:

[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 8, Time elapsed: 60.854 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 8, Time elapsed: 54.102 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 4, Time elapsed: 154.546 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 8, Time elapsed: 56.705 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 9, Time elapsed: 45.134 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 9, Time elapsed: 43.876 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
[ERROR] Tests run: 11, Failures: 0, Errors: 1, Skipped: 8, Time elapsed: 57.373 s <<< FAILURE! - in org.apache.ratis.grpc.TestRaftWithGrpc
...

CI:
https://github.com/adoroszlai/ratis/actions/runs/12445569381