End the check of a pull request's old head when a new head is checked

A check judges one head; once the pull request has another, its verdict
answers nobody and its run holds a build machine the new head waits
behind. Asking a check of R#N at H now drops every waiting check of R#N
at another head and kills every one in flight, saying each in the log.
Announcements are unordered and a recheck asks a recorded head, so this
happens only when the forge confirms H is the pull request's current
head; otherwise nothing ends and that is said. A one-shot recheck ends
them within the call. Tries, walk builds, group checks, other pull
requests' checks and rechecks of the same head are left alone.

novox/hq issue 726

cmd/mesh-controller/upgrades.go

@@ -23,7 +23,12 @@ import (
 // **It holds the stores, not a copy of the decision.** What to do about an upgrade is read when
 // one arrives, so changing it takes effect on the next upgrade rather than on the next restart of
 // the control plane.
-type following struct{ open *stores }
+type following struct {
+	open *stores
+	// oneShot is a process that ends after the call (a recheck through `delivery check`): what it would do beside
+	// the call it does within it, or the process's exit would cut it off (novox/hq issue 726).
+	oneShot bool
+}
 
 // Upgraded sends the machines running a module the version the catalogue now considers current —
 // or records that they are behind, which is the default and needs no record.