Mixing of bool + float covariates in sport.region.MaxPHeuristic not allowed

[twallema]

I'm trying to perform a spatial clustering using spot.region.maxPHeuristic. As I mixed attributes of type bool and float64,

biome_Amazônia             bool
biome_Caatinga             bool
biome_Cerrado              bool
biome_Mata Atlântica       bool
biome_Pampa                bool
biome_Pantanal             bool
cx                      float64
cy                      float64
dtype: object

I got the following error stack,

Traceback (most recent call last):
  File "/Users/twa27/Documents/github/DENV-serotype-imputation/scripts/clustering/find-clusters.py", line 104, in <module>
    model.solve()
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/spopt/region/maxp.py", line 836, in solve
    max_p, label = maxp(
                   ^^^^^
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/spopt/region/maxp.py", line 97, in maxp
    distance_matrix = squareform(pdist(attr, metric="cityblock"))
                                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/scipy/spatial/distance.py", line 2300, in pdist
    return xpx.lazy_apply(_np_pdist, X, out,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/scipy/_lib/array_api_extra/_lib/_lazy.py", line 300, in lazy_apply
    out = wrapped(*args, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/scipy/_lib/array_api_extra/_lib/_lazy.py", line 350, in wrapper
    out = func(*args_list, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/twa27/miniforge3/envs/DENV-SEROTYPE-IMPUTATION/lib/python3.12/site-packages/scipy/spatial/distance.py", line 2338, in _np_pdist
    return pdist_fn(X, out=out, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: Unsupported dtype object

Running the analysis with covariates biome_Amazônia -> biome_Pantanal (exclusively type bool) works fine. Converting the bool covariates to float before feeding them to maxPHeuristic solved the error. I'm assuming internally the mix of non-numeric + numeric gets converted to a numpy array of type "object" which causes this function to misbehave.

Suggested fix: Input check on types of covariate columns.