feat(forecast): Rust forecast crate, bit-for-bit with the Python serving path

Phase 2a of docs/RUST_REWRITE.md. rust/forecast (ping-forecast) ports the
serving half of src/ml: the HII mirror gap fill, the hourly grid, every
feature (pandas' Kahan rolling sums included), the Mae Ngat dam frame,
HistGradientBoosting tree inference, the heuristic fallback, stale-threshold
handling and the P.1 flood-stage probabilities.

Hand-over from Python (src/ml/export.py):
- models/flood_<station>.json.gz next to each joblib bundle; train_all() writes
  it, retrain.sh promotes it, `python -m src.ml.export` exports existing ones.
- rust/forecast/flood_config.json: thresholds, stages, leads and notify
  constants, compiled in. tests/test_model_export.py fails while it differs
  from the Python source.

Parity (rust/forecast/tests/parity.rs, fixtures from
testdata/gen_forecast_fixtures.py on the production code path): four
scenarios incl. the 2024 flood with RID lagging behind HII. Merged readings,
all features and all levels identical; with all 15 local models 377/384
probabilities bit-identical, the rest within 2 ulp (exp).

Python refactor without behaviour change: predict.forecasts_from_long, and
get_latest_forecasts/load_latest/serving_frame take now/rain for
reproducible comparisons; the HII query orders its rows.

rust/forecast/src/rain.rs

@@ -0,0 +1,27 @@
+//! Catchment rain series (`src/ml/rain.py`: `catchment_mean`).
+
+use chrono::NaiveDateTime;
+use std::collections::BTreeMap;
+
+/// Hourly catchment-average precipitation (mm); NaN = no value.
+pub type RainSeries = BTreeMap<NaiveDateTime, f64>;
+
+/// pandas `mean(axis=1)` over the catchment points: the points that have a
+/// value, summed in column order, divided by their count.
+pub fn catchment_mean(points: &[Option<f64>]) -> Option<f64> {
+    let mut sum = 0.0;
+    let mut n = 0usize;
+    for v in points.iter().flatten() {
+        sum += *v;
+        n += 1;
+    }
+    (n > 0).then(|| sum / n as f64)
+}
+
+/// The series on the given hours (`Series.reindex`), NaN where absent.
+pub fn reindex(rain: &RainSeries, times: &[NaiveDateTime]) -> Vec<f64> {
+    times
+        .iter()
+        .map(|t| rain.get(t).copied().unwrap_or(f64::NAN))
+        .collect()
+}