Fix flaky BDD test - export cls dataset

[leoll2]

One of the BDD tests failed during a daily run. Investigate why (possibly a too strict timeout) and fix it.

Ref: https://github.com/open-edge-platform/training_extensions/actions/runs/25980011827/job/76366962408

  @export @classification
  Scenario Outline: Export classification project dataset -- @1.1              # tests/bdd/features/export_dataset.feature:51
    Given A classification project "animals" with labels ["cat", "dog"] exists # tests/bdd/steps/export_dataset_steps.py:23
    And the project contains the following image distribution:                 # tests/bdd/steps/export_dataset_steps.py:41
      | Label | Training | Validation | Testing |
      | cat   | 5        | 5          | 5       |
      | dog   | 5        | 5          | 5       |
    And the project contains the following video frame distribution:           # tests/bdd/steps/export_dataset_steps.py:133
      | Label | Training | Validation | Testing |
      | cat   | 3        | 1          | 1       |
      | dog   | 3        | 1          | 1       |
    When I export the project dataset in GETI format with filters={ }          # tests/bdd/steps/export_dataset_steps.py:224
      ASSERT FAILED: Expected job to be DONE, but got RUNNING, error: None
      Traceback (most recent call last):
        File "/home/runner/work/training_extensions/training_extensions/application/backend/.venv/lib/python3.13/site-packages/behave/model.py", line 1991, in run
          match.run(runner.context)
          ~~~~~~~~~^^^^^^^^^^^^^^^^
        File "/home/runner/work/training_extensions/training_extensions/application/backend/.venv/lib/python3.13/site-packages/behave/matchers.py", line 105, in run
          self.func(context, *args, **kwargs)
          ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "tests/bdd/steps/export_dataset_steps.py", line 228, in step_export_dataset
          job = export_dataset(str(context.base_url), str(project.id), export_format, filters)
        File "/home/runner/work/training_extensions/training_extensions/application/backend/tests/bdd/utils/jobs.py", line 47, in export_dataset
          return wait_for_job_completion(base_url, job.job_id)
        File "/home/runner/work/training_extensions/training_extensions/application/backend/tests/bdd/utils/jobs.py", line 31, in wait_for_job_completion
          assert job.status == JobStatus.DONE.name, f"Expected job to be DONE, but got {job.status}, error: {job.error}"
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      AssertionError: Expected job to be DONE, but got RUNNING, error: None