Failed runs when no GPU devices is present.

[lyskov]

I am getting the following error (using latest code from main branch). (note that previously it was possible to run it without GPU) Thanks,

WARNING: You are welcome to use the default MSA server, however keep in mind that it's a
limited shared resource only capable of processing a few thousand MSAs per day. Please
submit jobs only from a single IP address. We reserve the right to limit access to the
server case-by-case when usage exceeds fair use. If you require more MSAs: You can 
precompute all MSAs with `colabfold_search` or host your own API and pass it to `--host-url`

Traceback (most recent call last):
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 1281, in run
    jax.tools.colab_tpu.setup_tpu()
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/tools/colab_tpu.py", line 20, in setup_tpu
    raise RuntimeError(
RuntimeError: jax.tools.colab_tpu.setup_tpu() was required for older JAX versions running on older generations of TPUs, and should no longer be used.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 874, in backends
    backend = _init_backend(platform)
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 965, in _init_backend
    backend = registration.factory()
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 663, in factory
    return xla_client.make_c_api_client(plugin_name, updated_options, None)
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jaxlib/xla_client.py", line 199, in make_c_api_client
    return _xla.get_c_api_client(plugin_name, options, distributed_client)
jaxlib.xla_extension.XlaRuntimeError: FAILED_PRECONDITION: No visible GPU devices.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/local/localcolabfold/colabfold-conda/bin/colabfold_batch", line 8, in <module>
    sys.exit(main())
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 2046, in main
    run(
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 1286, in run
    if jax.local_devices()[0].platform == 'cpu':
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 1135, in local_devices
    process_index = get_backend(backend).process_index()
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 1011, in get_backend
    return _get_backend_uncached(platform)
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 990, in _get_backend_uncached
    bs = backends()
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/jax/_src/xla_bridge.py", line 890, in backends
    raise RuntimeError(err_msg)
RuntimeError: Unable to initialize backend 'cuda': FAILED_PRECONDITION: No visible GPU devices. (you may need to uninstall the failing plugin package, or set JAX_PLATFORMS=cpu to skip this backend.)
Traceback (most recent call last):
  File "/usr/local/localcolabfold/colabfold-conda/bin/rosie", line 4, in <module>
    __import__('pkg_resources').run_script('rosie==0.0.1', 'rosie')
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/pkg_resources/__init__.py", line 752, in run_script
    self.require(requires)[0].run_script(script_name, ns)
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/pkg_resources/__init__.py", line 1729, in run_script
    exec(script_code, namespace, namespace)
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/rosie-0.0.1-py3.10.egg/EGG-INFO/scripts/rosie", line 6, in <module>
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/rosie-0.0.1-py3.10.egg/rosie/__init__.py", line 79, in main
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/rosie-0.0.1-py3.10.egg/rosie/__init__.py", line 50, in execute_flag_file
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/site-packages/rosie-0.0.1-py3.10.egg/rosie/alpha-red.py", line 23, in colabfold
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/subprocess.py", line 421, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
  File "/usr/local/localcolabfold/colabfold-conda/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command 'colabfold_batch ./alpha-fold.input ./alpha-fold.output --num-models 1 --model-type alphafold2_multimer_v3' returned non-zero exit status 1.