【BUG】Connect runtime out of memory

**BUG REPORT**

1. Please describe the issue you observed:

- What did you do (The steps to reproduce)?

- What is expected to see?
task status is RUNNING
- What did you see instead?
task status is FAILED, and reason is OutOfMemory.
2. Please tell us about your environment:

3. Other information (e.g. detailed explanation, logs, related issues, suggestions on how to fix, etc):
a. run `jmap -heap 16 ` found memory top 1 is org.apache.rocketmq.connect.runtime.stats.ConnectStatsService.
b. use arthas watch sourceTaskTimesList and sinkTaskTimesList

**SOLUTION:**
```
private void sampling() {
this.lockSampling.lock();
try {
this.sourceTaskTimesList.add(new CallSnapshot(System.currentTimeMillis(), sourceTaskTimesTotal()));
if (this.sourceTaskTimesList.size() > (MAX_RECORDS_OF_SAMPLING + 1)) {
this.sourceTaskTimesList.removeFirst();
}
// !!! sourceTaskTimesList is wrong !!!
// !!! should be sinkTaskTimesList !!!
this.sinkTaskTimesList.add(new CallSnapshot(System.currentTimeMillis(), sinkTaskTimesTotal()));
if (this.sourceTaskTimesList.size() > (MAX_RECORDS_OF_SAMPLING + 1)) {
this.sourceTaskTimesList.removeFirst();
}
} finally {
this.lockSampling.unlock();
}
}
```

Start by locating ConnectStatsService and its sampling() method, then inspect how sourceTaskTimesList and sinkTaskTimesList are bounded. Verify the sampling logic against the reported OutOfMemory behavior and the code shown in the issue. Done means both task-history lists retain only the intended maximum records and the relevant runtime tests pass.

Written by the indexing model from the issue text.