Export API to take in example inputs and calibration inputs

🚀 The feature, motivation and pitch

Today we have a single arg example_inputs in the export() api, the proposal here is to add another param called calibration_inputs that is intended for static quantization, example_inputs will be only used for in the torch.export.export stage.

https://github.com/pytorch/executorch/blob/main/export/export.py#L429

def export(
    model: Union[nn.Module, Dict[str, nn.Module]],
    example_inputs: Union[
        List[tuple[torch.Tensor, ...]], Dict[str, List[tuple[torch.Tensor, ...]]]
    ],
    export_recipe: ExportRecipe,
    name: Optional[str] = None,
    dynamic_shapes: Optional[Union[Any, Dict[str, Any]]] = None,
    constant_methods: Optional[Union[Dict[str, Callable]]] = None,
    artifact_dir: Optional[str] = None,
) -> "ExportSession":

In the context of recipes, if user passed a recipe that requires calibration data but instead provided only example inputs, we will error out.

CC: @kimishpatel, @metascroy, @cbilgin

Alternatives

No response

Additional context

No response

RFC (Optional)

No response