unbreak mypy torch/quantization

Summary:

Somehow `mypy torch/quantization` got broken in the past couple of days:
https://gist.github.com/vkuzo/07af454246f0a68e6fa8929beeec7e0d
.  I didn't see any relevant PRs other than
#47725, which doesn't seem
related. The error doesn't seem real, as the arguments to
`_cudnn_rnn_flatten_weight` seem correct. For now,
ignoring the failure so we have a clean `mypy` run on
`torch/quantization`.

Test Plan:

```
mypy torch/quantization
```

Reviewers:

Subscribers:

Tasks:

Tags:

[ghstack-poisoned]

torch/nn/modules/rnn.py

@@ -172,8 +172,8 @@ def flatten_parameters(self) -> None:
                     torch._cudnn_rnn_flatten_weight(
                         self._flat_weights, num_weights,
                         self.input_size, rnn.get_cudnn_mode(self.mode),
-                        self.hidden_size, self.proj_size, self.num_layers,
-                        self.batch_first, bool(self.bidirectional))
+                        self.hidden_size, self.proj_size, self.num_layers,  # type: ignore
+                        self.batch_first, bool(self.bidirectional))  # type: ignore
 
     def _apply(self, fn):
         ret = super(RNNBase, self)._apply(fn)