[Java] Ugly formatting for CJK data containing full-width characters

[DPUkyle]

The formatted output of a feature containing Japanese text, or the output of --i18n for Chinese, Japanese or Korean languages looks ugly - the spacing is difficult to manage for the table's pipe delimiters.

I'll attach a text example as an image to avoid any font/display issues:

I already have a fix for this and will submit a PR soon.