Rebuild: the coordinator spools replies without inflating them

On homelab (1 Gbit, 20 connections, ~630k articles/s) the coordinator
used 1.3-1.7 cores, mostly GC of the text it inflated only to know a
compressed reply was complete, plus a copy of every reply. A raw reply
is now complete when it ends in ".\r\n" (no copy): 0.4-0.5 cores at
the same rate.

".\r\n" can also end a network read inside the compressed data (~1 in
16M reads, a few dozen times over the whole target). Such a chunk
doesn't inflate on the worker: it reports it (POST ranges/:id/corrupt),
the spooled chunk is removed and the range goes back to the queue; the
prefetcher fetches it again. The rest of that reply breaks the
connection's next command, so the connection is replaced.

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

app/services/spool.ts

@@ -53,6 +53,11 @@ class Spool {
     return readFile(this.path(rangeId, span))
   }
 
+  /** One chunk, e.g. one that doesn't inflate (a false end of reply; see XoverClient.compressedRaw). */
+  async removeChunk(rangeId: number, span: ChunkSpan) {
+    await rm(this.path(rangeId, span), { force: true })
+  }
+
   async remove(rangeId: number) {
     await rm(join(this.dir, String(rangeId)), { recursive: true, force: true })
   }