Improving documentation of edge cases and worthy notes

[binaryaaron]

Per discussion on #30, I'd like to use this issue to both

• collect cases that deserve more up-front documentation
• propose solutions for how to show those cases to users

Simply, we can add an extra page in the sphinx docs and reference it from the home page?

[jrmontag]

Adding to this list of examples. Following a conversation with @fionapigott today, I think it will be valuable to make a space for a thorough, guided tour for the user through the variations (and recommendations) in working with the "text" of a Tweet. This could possibly be in the form of a wiki on the repo, or additional docs as mentioned above.

Some specific examples of ideas that we discussed (no decisions were made, just surfacing things):

• consider a unique (or more obvious), intentional delimiter for joining the pieces in get_all_text()
• following some additional work integrating the outputs of the parser into a modeling pipeline, consider alternative ways of providing the end-user straightforward access to the "textual" pieces as defined in the tweet_text module