Fit logistic regression models predicting probability of revert including features for:

• is newcomer
• ORES is deployed
• is anon
• ... (all the other features we have in the models)

To do this we have to:

1. Create automated badwords lists for the large-enough wikis that don't adopt ores for rcfilters.
2. Build datasets including the ores features for each wiki that did or didn't adopt oress for rcfilters.
3. Add features for newcomer, ores being deployed, and is_anon
4. fit diff-in-diff models combining these features.