Compat: a copy's own listeners aren't ours to bind

The first run against INBUXA's store failed all eight tests identically,
before reading a single record: the copy carries that server's listeners on
25, 443, 465, 587, 110, 143, 993 and 995, and nothing in a test run is
root, so each one failed with "Permission denied (os error 13)".

The builder now remembers the listeners it adds, and under NO_INSERT drops
build errors for any it didn't. Every other error still stands, including a
bind failing on one of its own, so this can't hide the case where the
harness's own port is taken.

The copy isn't edited for this: its listeners are simply not what a compat
run needs, and it reaches the server over the compat- ones instead. Checked
that a NO_INSERT run still boots and that scim_tests, which takes the
ordinary path, still passes.

docs/spec/compat-tests.md

@@ -99,6 +99,16 @@ copy of INBUXA's data any more. The script copies from the source for each
 test and removes the copy afterwards, so the source stays clean; take it
 from a stopped server or a snapshot, never from under a running one.
 
+**Why the copy's own listeners are ignored.** A real server listens on 25,
+443, 993 and the rest. Nothing in a test run is root, so every one of those
+fails to bind and the run died before reading any data, with eight
+`Permission denied (os error 13)` errors and nothing about what to do.
+Under `NO_INSERT` the harness now drops build errors belonging to listeners
+it did not add itself, and keeps every other error, including a bind that
+fails on one of its own (`tests/src/utils/server.rs`). The copy is not
+edited to achieve this; its listeners are simply not what a compat run
+needs.
+
 **Why `compat-` listeners appear in the copy.** The harness needs listeners
 on its own ports, and the registry keys listeners by name. A real server
 has its own, and a production listener called `jmap` or `imap` collided