WIP: Support completion for standard emoji

[Tak]

I took a stab at adding standard unicode emoji to the emoji completion results.
I couldn't find emoji/shortcodes exposed anywhere in the public android api, so I grabbed a json file from the gemoji project (generated from the official unicode document)

There are some challenges here:

• We need to remember to update these as new versions get released
• All the shortcodes are in english, so users will get a different experience than with gboard (e.g. :pain: != 🍞)
  -- Maybe set up a translatable resource for the shortcodes?

There's also room for improvement:

• The emoji db could be a more "native" format than a big json blob
• We could reformat the db for our data access needs
• We could strip superfluous information (e.g. ios version) from the db
  -- These all mean that we'd need our own postprocessor that we would ideally document and version somewhere…

[a1batross]

The emoji db could be a more "native" format than a big json blob

When I did emoji reactions, I just generated *.java file with all supported by backend emojis. I think, this JSON blob also could be converted to just Java class with unneeded data stripped.

[Tak]

Applies to #1956 , I guess

[connyduck]

My thoughts on this:

• We need to automate the generaton of the emoji data somehow, manual updating would be a lot of work
• A Java/Kotlin file is probably best for storing the emoji information
• How to handle the emoji support in different Android versions? Not all versions support all emojis

Maybe set up a translatable resource for the shortcodes?

I think leaving them in English is acceptable for a first version, but they should be translated eventually. Translating like other strings could work, yes.

While the feature would be nice to have, I don't think it is worth the effort.

Fyi, I occasionally get asked by users how to input standard emoji in Tusky, and all had no problem to use their keyboard after I told them how to.

[Tak]

Ok, that's fair