Upstream metatomic TorchSim model

[CompRhys]

Move the torchsim model upstream.

[CompRhys]

this is the only bit that I am sus on how to correctly load a model here.

[CompRhys]

        # Load the model
        if isinstance(model, (str, bytes, pathlib.PurePath)):
            if not os.path.exists(model):
                raise InputError(f"given model path '{model}' does not exist")

            # only store the model in self.parameters if is it the path to a file
            self.parameters["model"] = str(model)

            model = load_atomistic_model(
                model, extensions_directory=extensions_directory
            )

        elif isinstance(model, torch.jit.RecursiveScriptModule):
            if model.original_name != "AtomisticModel":
                raise InputError(
                    "torch model must be 'AtomisticModel', "
                    f"got '{model.original_name}' instead"
                )
        elif isinstance(model, AtomisticModel):
            # nothing to do
            pass
        else:
            raise TypeError(f"unknown type for model: {type(model)}")

This is how models are loaded on the ase side

[Luthaf]

Thanks @CompRhys! We had a chat about this 1h ago, and plan to create a separate metatomic-torchsim (and metatomic-ase, …) python package that would live in this repository but have a separate version number and release schedule. This might take us a couple of weeks to do, how urgent is getting this code out of torchsim for you?

@HaoZeke in the mean time can you help bringing this PR in a state where it supports all the same features as the other engine integrations?

[HaoZeke]

Thanks @CompRhys! We had a chat about this 1h ago, and plan to create a separate metatomic-torchsim (and metatomic-ase, …) python package that would live in this repository but have a separate version number and release schedule. This might take us a couple of weeks to do, how urgent is getting this code out of torchsim for you?

In the same vein, @CompRhys, would you be OK with me making changes to this PR? I'd like to setup the python package first.

@HaoZeke in the mean time can you help bringing this PR in a state where it supports all the same features as the other engine integrations?

Yup, in the package 😁

[CompRhys]

Not urgent. We just cut a 0.5.2 release and so will be a while till we cut 0.5.3. Hopefully this could land in that release. The big switch also relies on there being new releases with the upstreamed interfaces for orb-models and sevenn.

Please feel free to take this over or delete!