[GCP DWS] Fix None issue when no provision timeout is provided

[Michaelvll]

It is possible that the provision_timeout is set to None in our cluster yaml, which will cause a wrong CLI to wait for MIG.

To reproduce:

experimental:
  config_overrides:
    gcp:
      managed_instance_group:
        # Setup the DWS config
        run_duration: 36000
        # Failover to other regions after 30 minutes.
        provision_timeout: 1800
resources:
  accelerators: H100:8
  cpus: 32+

Tested (run the relevant ones):

• Code formatting: bash format.sh
• Any manual or new tests for this PR (please specify below)
• All smoke tests: pytest tests/test_smoke.py
• Relevant individual smoke tests: pytest tests/test_smoke.py::test_fill_in_the_name
• Backward compatibility tests: conda deactivate; bash -i tests/backward_compatibility_tests.sh