[SPARK-47181][CORE][TESTS] Fix MasterSuite to validate the number o…

…f registered workers

### What changes were proposed in this pull request?

This PR aims to fix `MasterSuite` to validate the number of registered workers during `SPARK-46881: scheduling with workerSelectionPolicy *` tests.

### Why are the changes needed?

To fix a flakiness.
- https://github.com/apache/spark/actions/runs/8042308713/job/21962794853#step:10:17224

```
[info] - SPARK-46881: scheduling with workerSelectionPolicy - CORES_FREE_DESC (false) *** FAILED *** (178 milliseconds)
[info]   List("10004") did not equal List("10005") (MasterSuite.scala:728)
[info]   Analysis:
[info]   List(0: "10004" -> "10005")
```

### Does this PR introduce _any_ user-facing change?

No.

### How was this patch tested?

Pass the CIs.

### Was this patch authored or co-authored using generative AI tooling?

No.

Closes apache#45274 from dongjoon-hyun/SPARK-47181.

Authored-by: Dongjoon Hyun <[email protected]>
Signed-off-by: Dongjoon Hyun <[email protected]>

core/src/test/scala/org/apache/spark/deploy/master/MasterSuite.scala

@@ -705,7 +705,7 @@ class MasterSuite extends SparkFunSuite
 
       // Use different core and memory values to simplify the tests
       MockWorker.counter.set(10000)
-      (1 to 5).map { idx =>
+      (1 to 5).foreach { idx =>
         val worker = new MockWorker(master.self, conf)
         worker.rpcEnv.setupEndpoint(s"worker-$idx", worker)
         val workerReg = RegisterWorker(
@@ -718,7 +718,9 @@ class MasterSuite extends SparkFunSuite
           "http://localhost:8080",
           RpcAddress("localhost", 10000))
         master.self.send(workerReg)
-        worker
+        eventually(timeout(10.seconds)) {
+          assert(master.self.askSync[MasterStateResponse](RequestMasterState).workers.size === idx)
+        }
       }
 
       // An application with two executors