Allow the onboarding script to perform language identification

[benjaminking]

At a lower priority than all the other sub-tasks, the onboarding script should perform language identification at both the project and book level for each project. This is to help catch common configuration errors that exist in Paratext projects:

• language codes not set correctly, perhaps due to copying the settings from a different project
• different languages used for different books

We don't expect that the language identifier will know the minority languages, but it should output unknown for a language it doesn't recognize, rather than reporting the closest language that it does know.