rna-transcription: make object-based and add starter implementation

[FridaTveit]

Changed rna-transcription to have an object based implementation (see issue #177) and added starter implementation as it is currently the second exercise in the new ordering (see issue #178).

[stkent]

Looks good! I'm not sure whether I prefer returning null vs throwing an unimplemented exception in the stubbed implementation - I feel like null could in general lead to a confusing error message when the tests are first run, and that the thrown exception route gives us more control over that messaging and experience. What do you think, Frida? I don't feel particularly strongly either way so I'm happy to go with whatever you decide, just let me know :)

[FridaTveit]

I didn't think about that, thanks for mentioning it! :) An exception is probably better. Do you think UnsupportedOperationException with message "Method has not been implemented yet" would be okay?

[stkent]

@FridaTveit that exception type sounds good to me, and to others!

[stkent]

Thanks, @FridaTveit!