[aten] embedding_bag_byte_rowwise_offsets_out (#49561)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/49561
Out variant for embedding_bag_byte_rowwise_offsets
Test Plan:
```MKL_NUM_THREADS=1 OMP_NUM_THREADS=1 numactl -m 0 -C 3 ./buck-out/opt/gen/caffe2/caffe2/fb/predictor/ptvsc2_predictor_bench --scripted_model=/data/users/ansha/tmp/adindexer/merge/traced_merge_dper_fixes.pt --p
t_inputs=/data/users/ansha/tmp/adindexer/merge/container_precomputation_bs1.pt --iters=30000 --warmup_iters=10000 --num_threads=1 --pred_net=/data/users/ansha/tmp/adindexer/precomputation_merge_net.pb --c2_inp
uts=/data/users/ansha/tmp/adindexer/merge/c2_inputs_precomputation_bs1.pb --c2_sigrid_transforms_opt=1 --c2_use_memonger=1 --c2_apply_nomnigraph_passes --c2_weights=/data/users/ansha/tmp/adindexer/merge/c2_weig
hts_precomputation.pb --pt_enable_static_runtime --pt_cleanup_activations=true --pt_enable_out_variant=true --compare_results --do_profile```
Check embedding_bag_byte_rowwise_offsets_out is called in perf
Before: 0.081438
After: 0.0783725
Reviewed By: supriyar, hlu1
Differential Revision: D25620718
fbshipit-source-id: 83d5d0dd2e1f60c46e6727f73d5d8b52661b6767