fix: correct always-true state check in DockerVM.stop()

The condition 'state != "stopped" or state != "exited"' is a tautology,
so the state check was a no-op and a stop request was sent even for a
container that had already exited.

_get_container_state() never returns "stopped" (only "running",
"paused" or "exited"), so the intended negation of the condition used in
_fix_permissions() requires 'and', not 'or' (De Morgan's law).

Added a regression test asserting no stop query is issued for an
already-exited container.

tests/compute/docker/test_docker_vm.py

@@ -1503,3 +1503,19 @@ async def test_read_console_output_with_binary_mode(vm):
     with asyncio_patch('gns3server.compute.docker.docker_vm.DockerVM.stop'):
         await vm._read_console_output(input_stream, output_stream)
         output_stream.feed_data.assert_called_once_with(b"test")
+
+
+async def test_stop_exited_container_no_stop_query(vm):
+
+    vm._ubridge_hypervisor = None
+    vm._fix_permissions = MagicMock()
+
+    with asyncio_patch("gns3server.compute.docker.DockerVM._get_container_state", return_value="exited"):
+        with asyncio_patch("gns3server.compute.docker.Docker.query") as mock_query:
+            vm._permissions_fixed = False
+            await vm.stop()
+            assert not any(
+                call.args[:2] == ("POST", "containers/e90e34656842/stop")
+                for call in mock_query.mock_calls
+            )
+    assert vm.status == "stopped"