feat(tools): count and summarize every match in the area tools

aircraft_in_area and vessels_in_area return at most 200 rows, so a client
asked which operators dominate an area, or the average altitude there, could
only estimate from a truncated sample. Both tools now take an optional
group_by (aircraft: operator, origin_country, type_code; ships: type,
destination) and stats (airborne altitude and speed for aircraft, speed for
ships). Each is computed over every match before rows are capped. Breakdowns
list the ten largest groups plus other and unknown counts, and the summary
names the three largest.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

src/tools/queries/maritime.test.mjs

@@ -224,3 +224,28 @@ test('an area search names its area to the server; a vessel lookup asks for ever
   assert.equal(lookup.has('lat'), false);
   assert.equal(lookup.has('radius_km'), false);
 });
+
+test('vessels in an area can be counted by destination with a speed summary', async () => {
+  const catalog = composeCatalog({
+    tools: coreTools,
+    services: { vessels: vessels() },
+  });
+  const result = await catalog.call('vessels_in_area', {
+    area: bay,
+    group_by: 'destination',
+    stats: true,
+  });
+  assert.equal(
+    result.summary,
+    '2 vessels in 20 km around 37.780, -122.400; most by destination: OAKLAND (1).',
+  );
+  assert.deepEqual(result.data.breakdown, {
+    field: 'destination',
+    groups: [{ value: 'OAKLAND', count: 1 }],
+    other_count: 0,
+    unknown_count: 1,
+  });
+  assert.deepEqual(result.data.stats, {
+    speed_mps: { count: 2, min: 6.2, median: 6.2, mean: 6.2, max: 6.2 },
+  });
+});