numpy.in1d removed

[LourensVeen]

Describe the bug
numpy has removed the numpy.in1d function in numpy 2.4.0.

PeTar and tupan use it, and they're downloaded codes so this really needs to be fixed upstream, and then we need to upgrade to a new version.

To Reproduce
This actually shows up when trying to build the docs after building PeTar and/or tupan, because Sphinx imports everything and then crashes on the downloaded code because numpy.in1d isn't found:

./src/amuse_tupan/src/tupan/tupan/analysis/glviewer.py:102:      [self.ps](http://self.ps/)[np.in1d([self.ps](http://self.ps/).id, ps.id)] = ps
./src/amuse_petar/src/PeTar/tools/analysis/bse.py:463:    sel_history=np.in1d(type_change.bid, bid_merge)
./src/amuse_petar/src/PeTar/tools/analysis/bse.py:470:    sel_merge_init = np.in1d(ubid_history_unique, ubid_merge)
./src/amuse_petar/src/PeTar/tools/get_object_snapshot.py:234:            sel1 = np.in1d(data_temp.p1.id, idlist)
./src/amuse_petar/src/PeTar/tools/get_object_snapshot.py:235:            sel2 = np.in1d(data_temp.p2.id, idlist)
./src/amuse_petar/src/PeTar/tools/get_object_snapshot.py:239:            sel = np.in1d(data_temp.bid, idlist)
./src/amuse_petar/src/PeTar/tools/get_object_snapshot.py:272:            sel = np.in1d(data_temp.id, idlist)

It looks like we may not actually be using these files, so the codes will probably work fine, but it would be nice if we could build the documentation.

Expected behavior
A successful documentation build.

Additional context
Thanks @elkogerville for the report!

[rieder]

Indeed we don't use those files directly. Could we let Sphinx ignore unused files like that? All it would need from ./src/amuse_(code)/ is the interface.py file, everything else in that directory should be ignored.