Develop Crowdin to Gitlab translations repo pipeline

Mullvad will be using crowdin for localizing Mullvad Browser. We need a process/pipeline for getting the first batch of strings and any new strings in the future from crowdin into our translations repo.