fix(event_analyzer): call 'kill_test' only once

We may get flooded with the CRITICAL events and each of them will
trigger call of "kill_test" method where we send SIGUSR2 signal to
python main thread.
At first, we should call this method only once.
At second, we will fall into endless loop of failures where we
catch following old python bug: https://bugs.python.org/issue24283

So, run "kill_test" method only once to avoid endless failure loop.

sdcm/sct_events/events_analyzer.py

@@ -51,9 +51,18 @@ def run(self) -> None:
                 except TestFailure:
                     self.kill_test(sys.exc_info())
 
-    def kill_test(self, backtrace_with_reason) -> None:
+    def kill_test(self, backtrace_with_reason, memo={}) -> None:  # pylint: disable=dangerous-default-value
         self.terminate()
         if tester := TestConfig().tester_obj():
+            if memo:
+                # NOTE: in some cases we may get flooded with the CRITICAL events.
+                #       And we should call 'kill_test' only once to avoid following:
+                #       - Long awaiting of the events handler closing in the tearDown stage.
+                #         It won't return until all the events are processed causing job timeout.
+                #       - Catching old python bug: https://bugs.python.org/issue24283
+                #       Also, in general, it makes no sense to kill test more than once.
+                return
+            memo["kill_test_has_run"] = True
             tester.kill_test(backtrace_with_reason)
         else:
             LOGGER.error("No test was registered using `TestConfig.set_tester_obj()', do not kill")