edge-bench-rs: file stems and subject remainders by hand, mimalloc

The suffix and remainder regexes were most of a new file's description
(333k on boneless 2026): now plain byte matching, with the regex versions
kept as the reference in a test of tricky names and subjects. mimalloc
for the many small strings. Boneless 2026: 360k -> 495k headers/s;
check.ts still finds the outputs identical on all three captures.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tools/edge-bench-rs/src/main.rs

@@ -18,6 +18,10 @@ use rustc_hash::FxHashMap;
 use std::io::{Read, Write};
 use std::time::Instant;
 
+// Many small strings (message-ids, names): the system allocator was a measurable share.
+#[global_allocator]
+static GLOBAL: mimalloc::MiMalloc = mimalloc::MiMalloc;
+
 fn inflate(raw: &[u8]) -> Vec<u8> {
     let mut out = Vec::with_capacity(raw.len() * 6);
     if raw.starts_with(&[0x1f, 0x8b]) {
@@ -200,6 +204,9 @@ fn main() {
         }
     }
     let total = inflate_s + index_s;
+    #[allow(static_mut_refs)]
+    let p = unsafe { indexer::PROF };
+    eprintln!("describe {:.2} s ({} calls), dates {:.2} s, aggregate {:.2} s, evict+cutoff {:.2} s, judge+encode {:.2} s", p[0] as f64 / 1e9, p[1], p[2] as f64 / 1e9, p[3] as f64 / 1e9, p[4] as f64 / 1e9, p[5] as f64 / 1e9);
     println!("{group} {start}–{end}: {} chunks, {} MB compressed, {} MB text", files.len(), compressed / 1_000_000, text_bytes / 1_000_000);
     println!(
         "{} headers in {:.2} s ({:.0}/s on one core; wall {:.2} s)",