The previous Unicode null character escaping logic was introducing
actual Unicode null character if there was an escaped Unicode null
character in the text.

We fix this issue by updating the regular expression used to capture
Unicode null characters.

Changelog: fixed
EE: true