AggregateGeneric fails with np.median

[ml31415]

import numpy as np
import numpy_groupies as npg

idx = np.arange(100).repeat(100)
d = np.random.random(10000)
npg.aggregate(idx, d, np.median)

And we get:

TypingError: Failed in nopython mode pipeline (step: nopython frontend)
Internal error at <numba.core.typeinfer.CallConstraint object at 0x7f30ed04a230>.
Failed in nopython mode pipeline (step: analyzing bytecode)
Use of unsupported opcode (DICT_MERGE) found

File "<__array_function__ internals>", line 179:
<source missing, REPL/exec in use?>

During: resolving callee type: type(CPUDispatcher(<function median at 0x7f3110dbe3b0>))
During: typing of call at /home/michael/.local/lib/python3.10/site-packages/numpy_groupies/aggregate_numba.py (247)

Enable logging at debug level for details.

File ".local/lib/python3.10/site-packages/numpy_groupies/aggregate_numba.py", line 247:
        def loop(sortidx, group_idx, a, ret):
            <source elided>
                    raise ValueError("one or more indices in group_idx are too large")
                ret[ri] = func(a_srt[start_idx:stop_idx])