ANSI technically doesn't have an escape code for LF, since it just uses the 0x0A character. The regex \xA is incorrect (it should be \x0A) and is being treated as the two-character sequence "xA". But then once we've fixed that, replacing 0x0A with \n in removeAnsiEscapeCodes is a no-op. Better to just remove the str.replace(LF, '\n') call.

Fixes #12608.

- add `editor.useTabStops` for tab/backspace behaviour
- add `editor.trimAutoWhitespace` for auto inserted whitespace cleanup

If you are prefer to use spaces instead of tabs - in most cases you
don't want to see tabs anywhere (copy paste issues with other files).

This fixes several workflows:

- Creating new Debug file
- Creating new Tasks file
- Showing global settings (this one is cached, so it shows for the first
  time only right indentation, if you will change it will show old
  indentation till restart)

Failed to execute 'btoa' on 'Window': The string to be encoded contains characters outside of the Latin1 range. (fixes #5004)