classify: stale docstring example outputs (reclassify, equal_interval) and five classifiers with no example

[brendancol]

Some docstring examples in xrspatial/classify.py print results that do not match what the functions return, and five public classifiers have no Examples section.

Wrong example output (printed array differs from the actual result for the given input)

• reclassify: the numpy, dask, and cupy blocks all print
  [[nan,1,1,1,1],[1,2,2,2,2],[2,2,2,2,2],[2,3,3,3,3]]
  but running the example produces
  [[nan,1,1,1,1],[1,1,1,1,1],[1,2,2,2,2],[2,3,3,3,nan]].
• equal_interval: the block prints
  [[nan,0,0,0,0],[0,0,0,0,1],[1,1,1,1,1],[1,2,2,2,2],[2,2,2,2,nan]]
  but running it produces
  [[nan,0,0,0,0],[0,1,1,1,1],[2,2,2,2,2],[3,3,3,3,4],[4,4,4,4,nan]].
• binary: the printed array uses np.nan where numpy reprs plain nan.

The test_classify.py fixtures confirm the functions behave correctly, so the docstring outputs are stale, not the code.

Missing Examples section

std_mean, head_tail_breaks, percentiles, maximum_breaks, and box_plot have Parameters, Returns, and References but no Examples block.

Scope

Documentation only (docstrings). No behavior change. Found via the documentation sweep on the classify module.

[brendancol]

Already resolved by #3516 (merged 2026-06-26). That PR fixed the stale reclassify/equal_interval outputs and the binary nan repr, and added Examples sections to std_mean, head_tail_breaks, percentiles, maximum_breaks, and box_plot.

The issue stayed open because #3516 referenced it as (#3506) in parentheses rather than with a Fixes #3506 keyword, so it never auto-closed.

Verified on current main by running each example: every docstring output now matches the function's actual return value, and all five previously-missing Examples sections are present. Closing as completed.