COMPAT: fix numpy 2.0 incompatibility

[martinfleis]

Closes #722

The issue probably won't pop-up for users given it is in the legacy implementation you would need to import directly.

Also adding numpy nightly to dev env.

[martinfleis]

The CI still did not pick up numpy 2 in dev env... I am afraid I'd need to manually install it in another step as it gets pulled by plenty of other deps before we get to the pip part. I will revert that and keep just that fix.

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 83.9%. Comparing base (e82d1be) to head (0a9a58f). Report is 1 commits behind head on main.

Additional details and impacted files

[](https://app.codecov.io/gh/pysal/libpysal/pull/723?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=pysal) ```diff @@ Coverage Diff @@ ## main #723 +/- ## ===================================== Coverage 83.9% 83.9% ===================================== Files 145 145 Lines 15583 15583 ===================================== Hits 13077 13077 Misses 2506 2506 ``` | [Files](https://app.codecov.io/gh/pysal/libpysal/pull/723?dropdown=coverage&src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=pysal) | Coverage Δ | | |---|---|---| | [libpysal/cg/voronoi.py](https://app.codecov.io/gh/pysal/libpysal/pull/723?src=pr&el=tree&filepath=libpysal%2Fcg%2Fvoronoi.py&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=pysal#diff-bGlicHlzYWwvY2cvdm9yb25vaS5weQ==) | `73.7% <100.0%> (ø)` | | | [libpysal/examples/remotes.py](https://app.codecov.io/gh/pysal/libpysal/pull/723?src=pr&el=tree&filepath=libpysal%2Fexamples%2Fremotes.py&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=pysal#diff-bGlicHlzYWwvZXhhbXBsZXMvcmVtb3Rlcy5weQ==) | `76.4% <100.0%> (ø)` | |

[jGaboardi]

Are we going to fix the failure in the PR or in a subsequent PR?

[martinfleis]

Are we going to fix the failure in the PR or in a subsequent PR?

I'd say so. I don't know why we see it, so it will require more time to figure out.