Move retry logic to import artifacts stage from ci file.

Background

Currently we will retry the import artifacts stage on a failure. This can be wasteful if there is only a single failure so we should consider moving this logic within download.py. See the last line of the job's CI yaml for the current retry logic.

import-artifacts:
  stage: import-artifacts
  dependencies:
    - load-scripts
    - lint
  variables:
    ARTIFACT_DIR: "${ARTIFACT_STORAGE}/import-artifacts"
  script:
    - '"${PIPELINE_REPO_DIR}/stages/import-artifacts/import-run.sh"'
  artifacts:
    when: always
    paths:
      - "${ARTIFACT_DIR}/"
    expire_in: 1 week
  retry: 1

Acceptance Criteria

• Import Artifacts job will be retried on system failures
• Any failures within the code will either cause a system exit, or a retry that is handled w/in the code.

Definition of Done

• Job no longer requires a retry: 1 in the CI yaml to allow for retries