twitter-text-conformance tests should be integrated

[ianozsvald]

I suspect that once the twitter conformance tests are integrated we'll see that many tests fail (e.g. wrt the issue raised here about unicode hashtags https://github.com/ianozsvald/twitter-text-python/issues/4 ): https://github.com/twitter/twitter-text-conformance

The steps would be to branch, read in the yaml (e.g. extract.yml), iterate over each entry and test to see if ttp gives the right answer. Subsequently we could fix the issues that are raised and/or integrate them into the main test suite.