feat(collector): forecast and ntfy in ping-collector; ThaiWater keeps alerts current at night

Phase 2b/2c of docs/RUST_REWRITE.md. With FORECAST_IN_COLLECTOR=true (shared
.env) ping-collector, after each cycle that brought newer RID or HII readings
and at least hourly: loads the last 336 h (RID + HII mirror fill), fetches the
rain forecast, runs every exported model (rust/forecast), stores
forecast_history and forecast_current, and sends the ntfy notifications.
The Python API then serves forecast_current as /forecast (computing in-process
only when it is older than FORECAST_MAX_AGE_SECONDS) and neither forecasts nor
notifies.

ntfy (collector/src/notify.rs): a port of src/notify.py with the same topics,
texts, hysteresis, capacity guard and notification_state keys, checked word
for word on 751 messages generated by Python (testdata/gen_notify_cases.py).
The change: for stations whose ThaiWater twin is the same sensor, a reading
newer than RID's drives the level alerts (marked as ThaiWater telemetry), and
the feed counts as stale only when neither source has a reading for 3 h. RID
alone lagging is reported once after RID_STALE_ALERT_HOURS (18). This ends the
nightly "gauge feed stale" / "recovered" pairs caused by RID's late reports.

Commands: `forecast [--json] [--save] [--now] [--rain-file]`,
`notify --dry-run`. scripts/forecast_parity.py compares Python and the
collector on the live database (same moment, same rain); locally 48/48 rows
identical. install_collector.sh: models export + that check on every install,
--cutover-forecast (refuses on a failed check) and --rollback-forecast.

Trees are packed into 24-byte nodes: 70 MB peak with all 15 models (was 140).
Forecast values are written to NUMERIC as shortest round-trip text, so they
round as psycopg2's do. DB test schemas get a counter (parallel tests collided
on Windows' coarse clock).

src/config.py

@@ -156,6 +156,17 @@ class Config:
         "yes",
     )
     FORECAST_POLL_SECONDS = int(os.getenv("FORECAST_POLL_SECONDS", "60"))
+    # true: ping-collector also forecasts (rust/forecast) and sends the ntfy
+    # notifications; this app serves its stored forecast (forecast_current)
+    # and does neither. Read by both services from the shared .env.
+    FORECAST_IN_COLLECTOR = os.getenv("FORECAST_IN_COLLECTOR", "false").lower() in (
+        "1",
+        "true",
+        "yes",
+    )
+    # A stored forecast older than this is not served; /forecast then computes
+    # in-process as before (the collector is down or has no models).
+    FORECAST_MAX_AGE_SECONDS = int(os.getenv("FORECAST_MAX_AGE_SECONDS", "7200"))
 
     # Logging settings
     LOG_LEVEL = os.getenv("LOG_LEVEL", "INFO")
@@ -306,6 +317,7 @@ class Config:
             "RID_RETRY_MINUTES": cls.RID_RETRY_MINUTES,
             "GAP_FILL_DAYS_ON_START": cls.GAP_FILL_DAYS_ON_START,
             "COLLECTION_ENABLED": cls.COLLECTION_ENABLED,
+            "FORECAST_IN_COLLECTOR": cls.FORECAST_IN_COLLECTOR,
             "LOG_LEVEL": cls.LOG_LEVEL,
             "LOG_FILE": cls.LOG_FILE,
             "DATA_RETENTION_DAYS": cls.DATA_RETENTION_DAYS,