How duplicate detection works
Each line is turned into a comparison key. With case-insensitive matching the key is lowercased using Unicode rules, and with trimming enabled the leading and trailing whitespace is removed. The tool walks through the lines from top to bottom and keeps a line only if its key has not been seen before. This takes a single pass, so even lists with hundreds of thousands of lines are processed almost instantly.
When trimming is on, the output lines are also trimmed so the result is consistent. When case-insensitive matching is on, the version of the line that appears first is the one kept, so you control which capitalization survives by ordering your list.
Line endings
Text copied from Windows applications uses CRLF line endings while macOS and Linux use LF. Both are recognized, so lines that differ only by an invisible carriage return are correctly treated as duplicates. The downloaded file uses LF endings, which every modern editor reads correctly.