Unfriendly failure on non-overlapping catalogs

[delucchi-cmu]

Bug report

If you take two catalogs that are totally non-overlapping and try to crossmatch them, LSDB fails in a way that would be confusing to end-users.

ValueError: cannot call `vectorize` on size 0 inputs unless `otypes` is set

with stack trace:

src/lsdb/catalog/catalog.py:214: in crossmatch
    ddf, ddf_map, alignment = crossmatch_catalog_data(
src/lsdb/dask/crossmatch_catalog_data.py:117: in crossmatch_catalog_data
    left_pixels, right_pixels = get_healpix_pixels_from_alignment(alignment)
src/lsdb/dask/merge_catalog_functions.py:212: in get_healpix_pixels_from_alignment
    left_pixels = make_pixel(
../../../.virtualenvs/demo/lib/python3.12/site-packages/numpy/lib/function_base.py:2372: in __call__
    return self._call_as_normal(*args, **kwargs)
../../../.virtualenvs/demo/lib/python3.12/site-packages/numpy/lib/function_base.py:2365: in _call_as_normal
    return self._vectorize_call(func=func, args=vargs)
../../../.virtualenvs/demo/lib/python3.12/site-packages/numpy/lib/function_base.py:2450: in _vectorize_call
    ufunc, otypes = self._get_ufunc_and_otypes(func=func, args=args)

Attaching a dataset suitable for unit tests (100 points randomly generated inside HealpixPixel(0,0), with some silly string identifier). The pixels and MOC are totally disjoint from the small_sky* catalogs.

disjoint.csv

I'm not sure where exactly the condition should be caught, and if we should provide an error, or simply an empty catalog.

Before submitting
Please check the following:

• I have described the situation in which the bug arose, including what code was executed, information about my environment, and any applicable data others will need to reproduce the problem.
• I have included available evidence of the unexpected behavior (including error messages, screenshots, and/or plots) as well as a description of what I expected instead.
• If I have a solution in mind, I have provided an explanation and/or pseudocode and/or task list.

[hombit]

I have the same issue with my pipeline, where I cross-match a sparse Gaia variable catalog with PS1 (paths on PSC):

import lsdb
from dask.distributed import Client
from lsdb.core.search.pixel_search import PixelSearch

search_filter = PixelSearch([(1, 33)])
ps1_otmo = lsdb.read_hats(
    "/ocean/projects/phy210048p/shared/hats/catalogs/ps1/ps1_otmo",
    margin_cache="/ocean/projects/phy210048p/shared/hats/catalogs/ps1/ps1_otmo_10arcs",
    columns=(
        ['objID', 'raMean', 'decMean']
    ),
    search_filter=search_filter,
)

gaia_var = lsdb.read_hats(
    "/ocean/projects/phy210048p/malanche/zubercal-filtering/hats/gaia_dr3_vcep",
    search_filter=search_filter,
)

result = gaia_var.crossmatch(
    ps1_otmo,
    radius_arcsec=1.0,
    suffixes=["", ""],
)