Suggestion should be shown when password matches user input

Right now, when the password matches a user input passed to `zxcvbn`, the score gets reduced to 0 but no relevant feedback is created. For example, if I make the password the same as the email address:

This feedback is not relevant to this situation. It would be much better to have a suggestion generated such as "Don't use personal information such as your name or email address in your password."

Start at the zxcvbn entry point that accepts user inputs and trace how a password matching one of them is scored and assigned feedback. Reproduce the email-address example, then verify that the result includes a suggestion about avoiding personal information rather than only the current irrelevant feedback.

Written by the indexing model from the issue text.