Prompt user to save their work when they click the "Back to {locale} {locale-code} Sim List" link

[liammulh]

In phetsims/qa#869, @solaolateju noted that it would be a good idea to prompt the user to save their work if they click the "Back to {Locale} {locale-code}" link. We already have plans to warn the user to save their work if they are about to navigate away from the page in #325, but as a temporary measure, I think it would be a good idea to prompt the user to save if they click the "Back to {Locale} {locale-code}" link.

We might also be able to prompt them if they click the "Select Locale" link in the nav if we check the URL to verify they are on the translation form page.

When we do #325, we'll want to remove these warnings because they will be redundant.

[liammulh]

We might also be able to prompt them if they click the "Select Locale" link in the nav if we check the URL to verify they are on the translation form page.

I'm not sure if this is worth the extra effort and complexity it would introduce. I think it could be done, but it would involve some hackery to make the links work. I think we'd need to add an onClick to the links and then programatically navigate the user if they click "OK" in the window.confirm. I will discuss this with JB.

[liammulh]

Since fc6c136 covers the original request, I'm going to mark this as ready for review.

[liammulh]

@jbphet and I discussed this. Since the window.confirm is basically "dumb" in the sense that it doesn't actually check whether the form is dirty with respect to the database or whether the form is dirty with respect to the saved translation in phetsims/babel, we don't want to include this. It will prompt the user even if they have just saved their translation. It might actually be confusing rather than helpful.

That being said, eventually we want to have a proper prompt that asks the user to save before they leave if the form is dirty with respect to the database and dirty with respect to the saved translation in phetsims/babel.

[liammulh]

There is another issue for this #325.