SetFit fails with Numpy >2.0.0

[chschroeder]

Bug description

The setfit multiclass example (examples/examplecode/setfit_multiclass_classification) fails on an installation with recent dependencies:

Traceback (most recent call last):
  File "/home/cschroeder/.pyenv/versions/3.9.18/lib/python3.9/runpy.py", line 197, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/cschroeder/.pyenv/versions/3.9.18/lib/python3.9/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/home/cschroeder/workspaces/small-text/examples/examplecode/setfit_multiclass_classification.py", line 106, in <module>
    main(num_iterations=args.num_iterations)
  File "/home/cschroeder/workspaces/small-text/examples/examplecode/setfit_multiclass_classification.py", line 51, in main
    indices_labeled = initialize_active_learner(active_learner, train.y)
  File "/home/cschroeder/workspaces/small-text/examples/examplecode/setfit_multiclass_classification.py", line 84, in initialize_active_learner
    active_learner.initialize(indices_initial)
  File "/home/cschroeder/.pyenv/versions/small-text-3.9/lib/python3.9/site-packages/small_text/active_learner.py", line 154, in initialize
    self._retrain(indices_validation=indices_validation)
  File "/home/cschroeder/.pyenv/versions/small-text-3.9/lib/python3.9/site-packages/small_text/active_learner.py", line 404, in _retrain
    self._clf.fit(dataset, **self.fit_kwargs)
  File "/home/cschroeder/.pyenv/versions/small-text-3.9/lib/python3.9/site-packages/small_text/integrations/transformers/classifiers/setfit.py", line 246, in fit
    return self._fit(sub_train, sub_valid, setfit_train_kwargs)
  File "/home/cschroeder/.pyenv/versions/small-text-3.9/lib/python3.9/site-packages/small_text/integrations/transformers/classifiers/setfit.py", line 261, in _fit
    seed = np.random.randint(np.iinfo(int_dtype()).max)
  File "numpy/random/mtrand.pyx", line 798, in numpy.random.mtrand.RandomState.randint
  File "numpy/random/_bounded_integers.pyx", line 1332, in numpy.random._bounded_integers._rand_int64
ValueError: high is out of bounds for int64

Downgrading numpy solves this.

Environment:

Python version: 3.9
small-text version: 2.0.0.dev1
small-text integrations (e.g., transformers):
PyTorch version (if applicable): 2.5.1
numpy: >2.0.0