[BUG] argsort diverges between cpu and gpu on complexes with NaN #4502

Describe the bug
An array with complex64 values containing NaNs sorts differently on gpu and cpu streams.

To Reproduce

import mlx.core as mx
NAN = float("nan")
a = mx.array([complex(1, NAN), 1 + 1j, 2 + 0j], mx.complex64)
mx.argsort(a, stream=mx.cpu) # -> [0 1 2]
mx.argsort(a, stream=mx.gpu) # -> [1 2 0]

GPU streams behave correctly (input is sorted, NaN last, result matches np.argsort)
CPU streams return [0,1,2,...n] without sorting.

Expected behavior
Expected outputs to match. Ideally they should behave as numpy et al.

Desktop (please complete the following information):

• OS Version: MacOS 27.0 RC
• Version : main (v0.32.2 + d9add9d)

Additional context
Hit in PR #4501
Issue 100% written by me, a human.