Avoid slicing text inside multi-byte characters

Right-to-left line reordering walked each atom's byte range and sliced the
line at every byte, panicking when a neutral multi-byte character such as a
guillemet preceded the first strong character. Walk characters instead.

FIND/FINDALL located the REPLACE keyword in a Unicode-uppercased copy and
sliced the original with that offset; uppercasing can change byte length
(dotless i, ligatures). Uppercase ASCII only, which keeps offsets aligned.
FILLET and CHAMFER inline R/D shorthands now test the original text before
slicing past the first character.

Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>

src/modules/draw/modify/fillet.rs

@@ -1514,7 +1514,7 @@ impl CadCommand for FilletCommand {
                     return Some(CmdResult::NeedPoint);
                 }
                 // "R 5.0" inline shorthand
-                if upper.starts_with('R') {
+                if t.starts_with(['r', 'R']) {
                     let body = t[1..].trim();
                     if let Some(v) = crate::entities::common::parse_typed_length(body) {
                         if v >= 0.0 {
@@ -2109,7 +2109,7 @@ impl CadCommand for ChamferCommand {
                     return Some(CmdResult::NeedPoint);
                 }
                 // "D 5.0" or "D 5.0 3.0" inline shorthand
-                if upper.starts_with('D') {
+                if t.starts_with(['d', 'D']) {
                     let body = t[1..].trim();
                     let parts: Vec<f64> = body
                         .split_whitespace()