HunspellColorize

Calls on initialization to respect the user's terminal locale. Handles multi-byte character sequences and buffer boundary split recovery using . Uses standard wide-character functions (, ) to accurately detect word boundaries across UTF-8 text without mangling non-ASCII characters.

on a linux distro like nixos /usr/share doesn't exist but XDG_DATA_DIRS is set to the location of hunspell files (if they are installed ofc) so we look for each directory in XDG_DATA_DIRS, check if theres an en_us dictionary in it, if not, fall back to previous /usr/share/... path tested this on my nixos and on an ubuntu docker image

This PR modifies the color scheme of HunspellColorize to make misspelled words appear in red in the terminal. Previously, the program used ANSI bold sequences (\033[1m / \033[22m) which some terminals do not display clearly. Changes made: Updated START and STOP macros: #define START "\033[31m" #define STOP "\033[0m" This improves visibility of misspelled words and provides a clearer visual cue for users reading text through the program. The Hunspell functionality remains unchanged, and the program still correctly identifies misspelled words using the same dictionaries. Tested on WSL/Ubuntu with English dictionary (en_US.aff/en_US.dic). No additional dependencies are required. Suggested Usage: echo "This is a simpel txt" less -R Now, misspelled words like "simpel" and "txt" appear in red.