[meziantou]

Adds MA0243 (Usage, Info, disabled by default), which reports invisible characters and special spaces written directly in string and character literals, and a code fix that replaces them with escape sequences.

Reported characters

• Control characters: U+0000–U+001F (including tab), U+007F, U+0080–U+009F. CR/LF are not reported (MA0101)
• Special spaces: U+00A0, U+1680, U+2000–U+200A, U+202F, U+205F, U+3000
• Zero-width / format characters: soft hyphen, ZWSP, ZWNJ, word joiner, BOM, Hangul fillers, braille blank, interlinear annotations…
• Bidirectional control characters (Trojan Source): U+061C, U+200E, U+200F, U+202A–U+202E, U+2066–U+2069
• Line/paragraph separators: U+2028, U+2029

The characters used by emoji sequences (ZWJ, variation selectors, tag characters) are not reported. The list is explicit rather than based on CharUnicodeInfo, so the result does not depend on the Unicode version of the runtime (Visual Studio vs. .NET SDK).

Scope

• Regular, verbatim, raw, interpolated and UTF-8 strings, and char literals. Only the text of interpolated strings is analyzed; nested literals in the holes are analyzed on their own.
• For multi-line raw strings, the indentation and the whitespace after the opening / before the closing delimiter are not reported, as they are not part of the value.
• One diagnostic per character, located on the character.

Code fix

• Uses \t, \0, \a, \b, \f, \v, \uXXXX or \UXXXXXXXX (fixed-width escapes, never \x).
• Verbatim strings (including $@"" and @""u8) are converted to regular strings.
• No fix for raw strings, nor for verbatim interpolated strings whose format clause contains \ or ".
• The whole literal is rewritten, so the fixes of all the characters of a literal are identical and merged by Fix All.

docs/comparison-with-other-analyzers.md lists Sonar S2479 as a similar rule.