Fixes #746: add emerging_hotspots() for time-series trend analysis (#890)

* Fixes #746: add emerging_hotspots() for time-series hot/cold spot trend analysis

Combines per-time-step Getis-Ord Gi* z-scores with the Mann-Kendall
trend test to classify each pixel into one of 17 trend categories
(new, consecutive, intensifying, persistent, diminishing, sporadic,
oscillating, historical — for both hot and cold spots).

Supports numpy, cupy, and dask+numpy backends.

* Add emerging hotspots user guide notebook (fills slot 8)

Walks through building a synthetic dataset, running the analysis,
understanding the 17 trend categories, visualising results with
matplotlib, inspecting per-pixel time series, boundary modes, and
Dask integration.

Author: brendancol

xrspatial/__init__.py

@@ -13,6 +13,7 @@
 from xrspatial.classify import quantile  # noqa
 from xrspatial.classify import reclassify  # noqa
 from xrspatial.curvature import curvature  # noqa
+from xrspatial.emerging_hotspots import emerging_hotspots  # noqa
 from xrspatial.focal import mean  # noqa
 from xrspatial.hillshade import hillshade  # noqa
 from xrspatial.mahalanobis import mahalanobis  # noqa