Fix UTF-8 slicing panics: use floor_char_boundary for all truncation

Byte-position truncation (&s[..s.len().min(N)]) panics when position
N lands inside a multi-byte character. Fixed in parser debug logging,
API error messages, oneshot response logging, and CLI agent display.

Also fixed tool dispatch permissions (removed global fallback).

Co-Authored-By: Proof of Concept <poc@bcachefs.org>

src/cli/agent.rs

@@ -136,7 +136,7 @@ pub fn cmd_digest_links(do_apply: bool) -> Result<(), String> {
         for (i, link) in links.iter().enumerate() {
             println!("  {:3}. {} → {}", i + 1, link.source, link.target);
             if !link.reason.is_empty() {
-                println!("       ({})", &link.reason[..link.reason.len().min(80)]);
+                println!("       ({})", &link.reason[..link.reason.floor_char_boundary(link.reason.len().min(80))]);
             }
         }
         println!("\nTo apply: poc-memory digest-links --apply");