fix: char-boundary-safe name truncation in cron_add tool

The name fallback used byte-index slice on the prompt,
which panics on multi-byte UTF-8 characters like Chinese.
Use is_char_boundary() to find a safe truncation point.

src/tools/cron.rs

@@ -131,7 +131,19 @@ impl Tool for CronAddTool {
         let name = args
             .get("name")
             .and_then(|v| v.as_str())
-            .unwrap_or(&prompt[..prompt.len().min(50)])
+            .unwrap_or_else(|| {
+                // char-boundary-safe truncation to 50 bytes
+                let limit = 50;
+                if prompt.len() <= limit {
+                    prompt.as_str()
+                } else {
+                    let mut end = limit;
+                    while !prompt.is_char_boundary(end) {
+                        end -= 1;
+                    }
+                    &prompt[..end]
+                }
+            })
             .to_string();
         let model = args
             .get("model")