feat: rewrite react-native-spokestack using bob builder

[timmywil]

• Now written in TypeScript, Swift, and Kotlin!
• Add a new .speak() method for easy playback of text, which makes use of existing players in spokestack-android and spokestack-ios. The docs explain this is only for simple playback. With anything more complicated, the user will need to handle playback themselves.
• Add an example project in this repo that uses the development source
• Adjust the API to return Promises for all methods
• Organize all configuration options into sections

What this still needs:

• [x] Finish auto-generated docs using typedoc
• [x] Unit tests (this only includes a basic test to double check everything is exported properly in JS). Unit tests in Swift and Kotlin could be added later, but for now everything is tested using the example app. As long as all of the methods are callable and all of the config options are passable, the functionality itself can be tested in the corresponding ios and android libraries (spokestack-ios and spokestack-android).
• [x] Support play event in Android

Would love @noelweichbrodt's feedback on the Swift part and @space-pope's feedback on the Kotlin part especially.

Feedback on the JS API is also welcome.

Viewing the diff is probably going to be difficult considering all the automated changes from using bob. I suggest looking at example/src/App.tsx for a working example, or even running the example app locally in Android Studio and/or XCode.

[timmywil]

@space-pope And here I assumed the RN config was kebab-case so that it could be passed directly to Android. camelCase is preferred in JS land tho.

[space-pope]

Sample app works as advertised on my test Pixel 1 XL. Thanks for including that!