fix(jmx): fix wait for jmx up

As we now verify ports being open from sct runner, we cannot use it for
jmx as it listens on localhost.

Fix by verifying JMX being up by verification of log message in
scylla-jmx service status.

fixes: #8183
(cherry picked from commit abc641b)

sdcm/cluster.py

@@ -1222,7 +1222,9 @@ def db_up(self):
         return self.is_port_used(port=self.CQL_PORT, service_name="scylla-server")
 
     def jmx_up(self):
-        return self.is_port_used(port=7199, service_name="scylla-jmx")
+        return self.remoter.run(f"{self.systemctl} is-active scylla-jmx.service && "
+                                f"{self.systemctl} status scylla-jmx.service | grep 'JMX is enabled to receive remote connections on port'",
+                                timeout=10, ignore_status=True).return_code == 0
 
     def cs_installed(self, cassandra_stress_bin=None):
         if cassandra_stress_bin is None: