GPU-enabled multiVI integration yaml issue

[wlason]

If the use_gpu is set to True in yaml, the script errors with the following message:

  File "[path]/panpipes/panpipes/python_scripts/batch_correct_multivi.py", line 248, in <module>
    mvi.train( **multivi_training_args, **multivi_training_plan)
  File "[path]/panpipes_gpu/lib/python3.10/site-packages/scvi/model/_multivi.py", line 338, in train
    runner = self._train_runner_cls(
  File "[path]/conda/skylake/envs/panpipes_gpu/lib/python3.10/site-packages/scvi/train/_trainrunner.py", line 82, in __init__
    self.trainer = self._trainer_cls(
  File "[path]/panpipes_gpu/lib/python3.10/site-packages/scvi/train/_trainer.py", line 171, in __init__
    super().__init__(
  File "[path]/panpipes_gpu/lib/python3.10/site-packages/lightning/pytorch/utilities/argparse.py", line 70, in insert_env_defaults
    return fn(self, **kwargs)
TypeError: Trainer.__init__() got an unexpected keyword argument 'use_gpu'