feat(observability): emit telemetry for NotifyListener warm + GridCache hit/miss

Band-warm failures in the Rust propagation_ready pipeline were only
surfaced via Logger.warning, so Prometheus had no way to see them.
The /weather GridCache hit/miss was likewise invisible while its
/map counterpart (ScoreCache) already reported cache ratio.

- NotifyListener emits [:microwaveprop, :propagation, :notify_listener, :warm]
  with %{ok, err} measurements and %{valid_time} metadata after every
  propagation_ready notification. Public handle_propagation_ready/1 so
  tests can exercise the warm+telemetry path without Postgrex.
- GridCache.fetch/1, fetch_bounds/2, and fetch_point/3 emit
  [:microwaveprop, :weather, :grid_cache, :lookup] with a :hit | :miss
  metadata key on every ETS lookup.

lib/microwaveprop/propagation/notify_listener.ex

@@ -84,7 +84,14 @@ defmodule Microwaveprop.Propagation.NotifyListener do
     end
   end
 
-  defp handle_propagation_ready(valid_time) do
+  @doc """
+  Warm `ScoreCache` for every band at `valid_time`, prune old cache
+  entries, and broadcast `propagation_updated` to PubSub. Public so
+  tests can exercise the warm+telemetry path without standing up the
+  Postgrex.Notifications subscriber.
+  """
+  @spec handle_propagation_ready(DateTime.t()) :: :ok
+  def handle_propagation_ready(valid_time) do
     {ok, err} =
       Enum.reduce(BandConfig.all_bands(), {0, 0}, fn band, {ok, err} ->
         case warm_band(band.freq_mhz, valid_time) do
@@ -93,6 +100,12 @@ defmodule Microwaveprop.Propagation.NotifyListener do
         end
       end)
 
+    :telemetry.execute(
+      [:microwaveprop, :propagation, :notify_listener, :warm],
+      %{ok: ok, err: err},
+      %{valid_time: valid_time}
+    )
+
     # Prune anything older than 2 hours so the cache doesn't grow
     # unbounded when Rust covers many forecast hours back-to-back.
     ScoreCache.prune_older_than(DateTime.add(DateTime.utc_now(), -2, :hour))
@@ -104,6 +117,8 @@ defmodule Microwaveprop.Propagation.NotifyListener do
     else
       Logger.info("NotifyListener: warmed ScoreCache + broadcast for valid_time=#{valid_time}")
     end
+
+    :ok
   end
 
   defp warm_band(band_mhz, valid_time) do

lib/microwaveprop/weather/grid_cache.ex

@@ -32,16 +32,26 @@ defmodule Microwaveprop.Weather.GridCache do
   @spec fetch(DateTime.t()) :: {:ok, [row()]} | :miss
   def fetch(valid_time) do
     case :ets.lookup(@table, valid_time) do
-      [{_, grid}] -> {:ok, grid_to_list(grid)}
-      [] -> :miss
+      [{_, grid}] ->
+        emit_lookup(true)
+        {:ok, grid_to_list(grid)}
+
+      [] ->
+        emit_lookup(false)
+        :miss
     end
   end
 
   @spec fetch_bounds(DateTime.t(), bounds() | nil) :: {:ok, [row()]} | :miss
   def fetch_bounds(valid_time, bounds) do
     case :ets.lookup(@table, valid_time) do
-      [{_, grid}] -> {:ok, grid_to_filtered_list(grid, bounds)}
-      [] -> :miss
+      [{_, grid}] ->
+        emit_lookup(true)
+        {:ok, grid_to_filtered_list(grid, bounds)}
+
+      [] ->
+        emit_lookup(false)
+        :miss
     end
   end
 
@@ -50,15 +60,25 @@ defmodule Microwaveprop.Weather.GridCache do
     case :ets.lookup(@table, valid_time) do
       [{_, grid}] ->
         case Map.get(grid, {lat, lon}) do
-          nil -> :miss
-          row -> {:ok, row}
+          nil ->
+            emit_lookup(false)
+            :miss
+
+          row ->
+            emit_lookup(true)
+            {:ok, row}
         end
 
       [] ->
+        emit_lookup(false)
         :miss
     end
   end
 
+  defp emit_lookup(hit) do
+    :telemetry.execute([:microwaveprop, :weather, :grid_cache, :lookup], %{}, %{hit: hit})
+  end
+
   @spec put(DateTime.t(), [row()]) :: :ok
   def put(valid_time, rows) do
     grid = list_to_grid(rows)

test/microwaveprop/propagation/notify_listener_test.exs

@@ -0,0 +1,114 @@
+defmodule Microwaveprop.Propagation.NotifyListenerTest do
+  # async: false — uses the shared :propagation_scores_dir + ScoreCache
+  use ExUnit.Case, async: false
+
+  alias Microwaveprop.Propagation.BandConfig
+  alias Microwaveprop.Propagation.Grid
+  alias Microwaveprop.Propagation.NotifyListener
+  alias Microwaveprop.Propagation.ScoreCache
+  alias Microwaveprop.Propagation.ScoresFile
+
+  setup do
+    dir =
+      Path.join(
+        System.tmp_dir!(),
+        "notify_listener_test_#{System.unique_integer([:positive])}"
+      )
+
+    File.mkdir_p!(dir)
+    prev = Application.get_env(:microwaveprop, :propagation_scores_dir)
+    Application.put_env(:microwaveprop, :propagation_scores_dir, dir)
+    ScoreCache.clear()
+
+    on_exit(fn ->
+      File.rm_rf!(dir)
+      ScoreCache.clear()
+
+      if prev do
+        Application.put_env(:microwaveprop, :propagation_scores_dir, prev)
+      else
+        Application.delete_env(:microwaveprop, :propagation_scores_dir)
+      end
+    end)
+
+    :ok
+  end
+
+  defp sample_scores do
+    %{lat_min: lat_min, lon_min: lon_min, lat_max: lat_max, lon_max: lon_max} = Grid.bounds()
+
+    [
+      %{lat: lat_min, lon: lon_min, score: 10},
+      %{lat: lat_max, lon: lon_max, score: 90}
+    ]
+  end
+
+  describe "handle_propagation_ready/1 telemetry" do
+    test "emits propagation.notify_listener.warm with ok/err counts after warming every band" do
+      valid_time = ~U[2026-04-21 12:00:00Z]
+
+      # Write a scores file for the first band only so the rest hit :enoent
+      # and contribute to the err tally. That gives us at least one ok and
+      # at least one err for an all-bands warm.
+      [first_band | _rest] = Enum.map(BandConfig.all_bands(), & &1.freq_mhz)
+      ScoresFile.write!(first_band, valid_time, sample_scores())
+
+      total_bands = length(BandConfig.all_bands())
+      expected_err = total_bands - 1
+
+      test_pid = self()
+      handler_id = {__MODULE__, :notify_listener_warm}
+
+      :telemetry.attach(
+        handler_id,
+        [:microwaveprop, :propagation, :notify_listener, :warm],
+        fn event, measurements, metadata, _config ->
+          send(test_pid, {:telemetry, event, measurements, metadata})
+        end,
+        nil
+      )
+
+      try do
+        :ok = NotifyListener.handle_propagation_ready(valid_time)
+
+        event = [:microwaveprop, :propagation, :notify_listener, :warm]
+
+        assert_receive {:telemetry, ^event, %{ok: 1, err: ^expected_err}, %{valid_time: ^valid_time}}
+      after
+        :telemetry.detach(handler_id)
+      end
+    end
+
+    test "emits telemetry even when every band warms successfully" do
+      valid_time = ~U[2026-04-21 13:00:00Z]
+
+      for band <- BandConfig.all_bands() do
+        ScoresFile.write!(band.freq_mhz, valid_time, sample_scores())
+      end
+
+      total_bands = length(BandConfig.all_bands())
+
+      test_pid = self()
+      handler_id = {__MODULE__, :notify_listener_warm_all_ok}
+
+      :telemetry.attach(
+        handler_id,
+        [:microwaveprop, :propagation, :notify_listener, :warm],
+        fn event, measurements, metadata, _config ->
+          send(test_pid, {:telemetry, event, measurements, metadata})
+        end,
+        nil
+      )
+
+      try do
+        :ok = NotifyListener.handle_propagation_ready(valid_time)
+
+        event = [:microwaveprop, :propagation, :notify_listener, :warm]
+
+        assert_receive {:telemetry, ^event, %{ok: ^total_bands, err: 0}, %{valid_time: ^valid_time}}
+      after
+        :telemetry.detach(handler_id)
+      end
+    end
+  end
+end

test/microwaveprop/weather/grid_cache_test.exs

@@ -121,6 +121,68 @@ defmodule Microwaveprop.Weather.GridCacheTest do
     end
   end
 
+  describe "lookup telemetry" do
+    @event [:microwaveprop, :weather, :grid_cache, :lookup]
+
+    setup do
+      test_pid = self()
+      handler_id = {__MODULE__, :grid_cache_lookup, System.unique_integer([:positive])}
+
+      :telemetry.attach(
+        handler_id,
+        @event,
+        fn event, measurements, metadata, _config ->
+          send(test_pid, {:telemetry, event, measurements, metadata})
+        end,
+        nil
+      )
+
+      on_exit(fn -> :telemetry.detach(handler_id) end)
+      :ok
+    end
+
+    test "fetch/1 emits hit: true when the valid_time is cached" do
+      GridCache.put(~U[2026-04-12 12:00:00Z], [%{lat: 32.0, lon: -97.0}])
+      {:ok, _} = GridCache.fetch(~U[2026-04-12 12:00:00Z])
+      assert_receive {:telemetry, @event, %{}, %{hit: true}}
+    end
+
+    test "fetch/1 emits hit: false when the valid_time is absent" do
+      :miss = GridCache.fetch(~U[2099-01-01 00:00:00Z])
+      assert_receive {:telemetry, @event, %{}, %{hit: false}}
+    end
+
+    test "fetch_bounds/2 emits hit: true on a cached valid_time" do
+      GridCache.put(~U[2026-04-12 12:00:00Z], [%{lat: 32.0, lon: -97.0}])
+      bounds = %{"south" => 0.0, "north" => 90.0, "west" => -180.0, "east" => 0.0}
+      {:ok, _} = GridCache.fetch_bounds(~U[2026-04-12 12:00:00Z], bounds)
+      assert_receive {:telemetry, @event, %{}, %{hit: true}}
+    end
+
+    test "fetch_bounds/2 emits hit: false when the valid_time is absent" do
+      bounds = %{"south" => 0.0, "north" => 90.0, "west" => -180.0, "east" => 0.0}
+      :miss = GridCache.fetch_bounds(~U[2099-01-01 00:00:00Z], bounds)
+      assert_receive {:telemetry, @event, %{}, %{hit: false}}
+    end
+
+    test "fetch_point/3 emits hit: true when the point exists" do
+      GridCache.put(~U[2026-04-12 12:00:00Z], [%{lat: 32.0, lon: -97.0}])
+      {:ok, _} = GridCache.fetch_point(~U[2026-04-12 12:00:00Z], 32.0, -97.0)
+      assert_receive {:telemetry, @event, %{}, %{hit: true}}
+    end
+
+    test "fetch_point/3 emits hit: false when the valid_time is missing" do
+      :miss = GridCache.fetch_point(~U[2099-01-01 00:00:00Z], 1.0, 2.0)
+      assert_receive {:telemetry, @event, %{}, %{hit: false}}
+    end
+
+    test "fetch_point/3 emits hit: false when the point is missing from a cached grid" do
+      GridCache.put(~U[2026-04-12 12:00:00Z], [%{lat: 32.0, lon: -97.0}])
+      :miss = GridCache.fetch_point(~U[2026-04-12 12:00:00Z], 40.0, -74.0)
+      assert_receive {:telemetry, @event, %{}, %{hit: false}}
+    end
+  end
+
   describe "latest_valid_time/0" do
     test "returns the most recent cached valid_time" do
       GridCache.put(~U[2026-04-12 10:00:00Z], [])