[5693592][ONNX-customOp][Autocast] Fix QuantizeLinear node output type (#671)

## What does this PR do?

**Type of change:** Bug fix

**Overview:** The output type of Q nodes was being set incorrectly. This
PR fixes that.

## Usage

```python
$ python -m modelopt.onnx.autocast --onnx_path=$MODEL_NAME.onnx
```

## Testing
See bug 5693592 for more details:
```python
$ python -m modelopt.onnx.autocast --onnx_path=$MODEL_NAME.onnx --low_precision_type=fp16 --data_max=inf --init_max=inf --keep_io_types
```

## Before your PR is "*Ready for review*"
<!-- If you haven't finished some of the above items you can still open
`Draft` PR. -->

- **Make sure you read and follow [Contributor
guidelines](https://github.com/NVIDIA/Model-Optimizer/blob/main/CONTRIBUTING.md)**
and your commits are signed.
- **Is this change backward compatible?**: Yes
- **Did you write any new necessary tests?**: No
- **Did you add or update any necessary documentation?**: No
- **Did you update
[Changelog](https://github.com/NVIDIA/Model-Optimizer/blob/main/CHANGELOG.rst)?**:
No

Signed-off-by: gcunhase <[email protected]>

Author: gcunhase

modelopt/onnx/autocast/precisionconverter.py

@@ -296,6 +296,8 @@ def _get_np_type(node, inp, opset=onnx.defs.onnx_opset_version()):
                 return helper.tensor_dtype_to_np_dtype(node.attrs["to"])
             elif node.op == "DequantizeLinear":
                 return node.inputs[1].dtype  # scale type
+            elif node.op == "QuantizeLinear":
+                return node.inputs[2].dtype  # zero_point type
             elif not inp.dtype or inp.dtype == onnx.TensorProto.UNDEFINED:
                 return None
             elif node.op not in self.custom_ops: