Improve barrier synchronization in ParallelBuildChainTest with shared timeout

Refactors the job synchronization logic to use a shared timeout when waiting
for all jobs to reach running state, rather than waiting sequentially for each
barrier. Adds a non-blocking getStatus() method to TestBarrier2 to support
this more efficient synchronization approach.

Fixes #825

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

Author: vogella

resources/tests/org.eclipse.core.tests.resources/src/org/eclipse/core/tests/internal/builders/ParallelBuildChainTest.java

@@ -447,9 +447,9 @@ protected IStatus run(IProgressMonitor monitor) {
 			job.setJobGroup(jobGroup);
 			job.schedule();
 		}
-		for (TestBarrier2 barrier : waitForRunningJobBarriers.values()) {
-			barrier.waitForStatus(TestBarrier2.STATUS_RUNNING);
-		}
+		// Wait for all job wrappers to reach running state with a shared timeout
+		waitForCondition(() -> waitForRunningJobBarriers.values().stream()
+				.allMatch(barrier -> barrier.getStatus() == TestBarrier2.STATUS_RUNNING), TIMEOUT_IN_MILLIS);
 		try {
 			executeWhileRunningBuild.run();
 		} finally {

runtime/tests/org.eclipse.core.tests.harness/src/org/eclipse/core/tests/harness/TestBarrier2.java

@@ -183,6 +183,13 @@ public TestBarrier2(AtomicIntegerArray location, int index) {
 		this.myIndex = index;
 	}
 
+	/**
+	 * Gets this barrier object's current status.
+	 */
+	public int getStatus() {
+		return myStatus.get(myIndex);
+	}
+
 	/**
 	 * Sets this barrier object's status.
 	 */