[Site] Next and Prev should Skip Pages not Containing Selected Dialects

[pishoyg]

Following up on #122, we've been using JavaScript for the dialected variants. The site should be smart enough to skip through pages that have neither a word nor a derivation belonging to any of the selected dialect.

You can do this by redirecting to the "next" link. Maybe also display a popup! Waiting for a second before doing it would also be nice!

#62 is somewhat related. We might be able to solve both by having the Crum pipeline export a piece of metadata, representing all dialects present in a word¹.

Footnotes

1. A word in the context of Crum refers to a dictionary entry, with all the spellings and derivations - that's a note in the Flashcards context, or a page in the Site context. ↩

[pishoyg]

NOTE: Handle the case when zero dialects are selected!
NOTE: Handle the case when undialected root spellings are present. (Undialected derivation spellings have no significance.)

[pishoyg]

There is no need to include any extra metadata in the website. Checking for the presence of an element bearing the dialect code should suffice.
But we might want to change the HTML slightly to include classes that distinguish between roots and derivations, in order to handle the case of undialected roots (which have to receive a different treatment from undialected derivations).

[pishoyg]

NOTE: We're removing these buttons from Anki! See.