fix: skip empty parameters in gradient reduction
Empty parameters (numel=0) cause issues in gradient allreduce when
using flatten/unflatten operations. The unflatten operation fails
with shape mismatches because empty tensors can't be properly
reconstructed from a flattened buffer.
This fix skips empty parameters in _get_gradients_for_reduction()
since they contribute nothing to gradient reduction anyway.
Fixes test_onebit.py::TestOneBitLambEmptyParameters::test
Signed-off-by: Masahiro Tanaka <mtanaka@anyscale.com>
