tests/partition_balancer: more robust wait_until_status

Previously, when the controller leader node was suspended during the
test all status requests would fail with the timed-out error.
This was true for all nodes, not just the suspended one (because we
proxy the status request to the controller leader), so internal retries
in the admin API wrapper didn't help. We increase the timeout and add
504 to retriable status codes so that internal retries can handle this
situation.

tests/rptest/tests/partition_balancer_test.py

@@ -77,13 +77,15 @@ def node2partition_count(self):
         return ret
 
     def wait_until_status(self, predicate, timeout_sec=120):
-        admin = Admin(self.redpanda)
+        # We may get a 504 if we proxy a status request to a suspended node.
+        # It is okay to retry (the controller leader will get re-elected in the meantime).
+        admin = Admin(self.redpanda, retry_codes=[503, 504])
         start = time.time()
 
         def check():
             req_start = time.time()
 
-            status = admin.get_partition_balancer_status(timeout=1)
+            status = admin.get_partition_balancer_status(timeout=10)
             self.logger.info(f"partition balancer status: {status}")
 
             if "seconds_since_last_tick" not in status: