fix(bandopen): the moon is not a band opening

A 2 m opening was announced at 9650 km towards Japan. The callsigns gave it
away — 7M4RRM, JA7RPC, JF1AWC — all working EME, which is routine on 2 m and
reported to PSK Reporter like anything else. Real contacts, real grids, and no
evidence whatsoever about the band: an operator turning a beam on that bearing
would hear nothing.

Removing the flat 2400 km ceiling was right; it threw away genuine multi-hop Es
on 6 m. "No limit anywhere" was the overcorrection. The limit is per band now,
and it is physics rather than a threshold: 2 m reaches a few thousand kilometres
by tropospheric duct or a chain of Es clouds and no further, so 3500 km, and
4 m 4000. Six and ten metres keep no ceiling — multi-hop Es and F2 genuinely do
go round the world, which is the case that started all this.

Pinned from both sides: the 9650 km Japanese burst produces nothing, and a
2000 km 2 m burst in one sector still counts.

internal/bandopen/bandopen.go

@@ -81,6 +81,28 @@ var watched = map[string]bool{"10m": true, "6m": true, "4m": true, "2m": true}
 // Watched reports whether a band is one the detector looks at.
 func Watched(band string) bool { return watched[strings.ToLower(strings.TrimSpace(band))] }
 
+// maxTerrestrialKm is the longest path a band can carry through the atmosphere.
+// Zero means no limit.
+//
+// The flat 2400 km ceiling was removed because it threw away real multi-hop Es
+// on 6 m, and that was right — but "no limit anywhere" then let something else
+// through. A 2 m opening was announced at 9650 km towards Japan, on stations
+// that were unmistakably working EME: the moon is not an opening, and pointing
+// an antenna at that bearing would find nothing.
+//
+// So the limit is per band, and it is physics rather than a threshold. Two
+// metres reaches a few thousand kilometres by tropospheric duct or a chain of Es
+// clouds and no further; beyond that the path went via the moon or a satellite,
+// neither of which says anything about the band. Six and ten metres have no
+// ceiling at all — multi-hop Es and F2 genuinely go round the world.
+var maxTerrestrialKm = map[string]int{
+	"2m": 3500,
+	"4m": 4000,
+}
+
+// MaxKmFor returns the plausibility ceiling for a band, 0 for none.
+func MaxKmFor(band string) int { return maxTerrestrialKm[strings.ToLower(strings.TrimSpace(band))] }
+
 // Opening is a detected opening, ready to be announced.
 type Opening struct {
 	Band       string `json:"band"`
@@ -130,6 +152,12 @@ func (d *Detector) Add(s Spot, lat float64) *Opening {
 	if s.DistKm < d.cfg.MinKm || (d.cfg.MaxKm > 0 && s.DistKm > d.cfg.MaxKm) {
 		return nil
 	}
+	// Past what the atmosphere can carry on this band, the path went via the moon
+	// or a satellite. Those are real contacts and real reports; they are simply
+	// not evidence about the band.
+	if m := MaxKmFor(band); m > 0 && s.DistKm > m {
+		return nil
+	}
 	d.recent = append(d.recent, s)
 	d.prune(s.At)