Examples use `mps` as accelerator

[tizianocitro]

Describe the bug

Example pipelines break on machines with no mps.

To reproduce

Steps to reproduce the behavior:

1. Run make run examples/hgnn.py on a machine with no mps.

Expected behavior

It should be able to run on CUDA directly.

Additional context

make run examples/hgnn.py produces:

=== Run examples/hgnn.py ===
uv run python3 examples/hgnn.py
Loading and preparing dataset...
Enriching node features...
Creating dataloaders...
Starting training and evaluation...
Traceback (most recent call last):
  File "/workspace/hyperbench/examples/hgnn.py", line 143, in <module>
    with MultiModelTrainer(
         ^^^^^^^^^^^^^^^^^^
  File "/workspace/hyperbench/hyperbench/train/trainer.py", line 174, in __init__
    model_config.trainer = L.Trainer(
                           ^^^^^^^^^^
  File "/workspace/hyperbench/.venv/lib/python3.12/site-packages/lightning/pytorch/utilities/argparse.py", line 70, in insert_env_defaults
    return fn(self, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/workspace/hyperbench/.venv/lib/python3.12/site-packages/lightning/pytorch/trainer/trainer.py", line 417, in __init__
    self._accelerator_connector = _AcceleratorConnector(
                                  ^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/hyperbench/.venv/lib/python3.12/site-packages/lightning/pytorch/trainer/connectors/accelerator_connector.py", line 145, in __init__
    self._set_parallel_devices_and_init_accelerator()
  File "/workspace/hyperbench/.venv/lib/python3.12/site-packages/lightning/pytorch/trainer/connectors/accelerator_connector.py", line 356, in _set_parallel_devices_and_init_accelerator
    raise MisconfigurationException(
lightning.fabric.utilities.exceptions.MisconfigurationException: `MPSAccelerator` can not run on your system since the accelerator is not available. The following accelerator(s) is available and can be passed into `accelerator` argument of `Trainer`: ['cpu', 'cuda'].