Fix repeated gradient scaling across pipeline stages

[Functionhx]

Summary

Fix repeated gradient accumulation scaling when pipeline parallelism is combined with gradient accumulation.

Every pipeline stage registers an output backward hook that scales gradients by gradient_accumulation_steps. The last stage therefore scales the gradient correctly before sending it upstream, but each preceding stage scales the already-scaled gradient again.

Fixes #8152

Changes

• Apply gradient accumulation scaling only at the last pipeline stage.
• Preserve the existing behavior for single-stage pipelines and gradient_accumulation_steps=1.
• Add a deterministic two-stage regression test that verifies both stages receive correctly averaged gradients.
• Disable gradient clipping in the regression test so it cannot mask the scaling difference.

Testing

• DS_ACCELERATOR=cpu LOCAL_SIZE=2 OMP_NUM_THREADS=1 pytest -q tests/unit/runtime/pipe/test_pipe.py::TestPipeGradientAccumulationScaling::test_gradients_are_scaled_once
• pre-commit run --files deepspeed/runtime/pipe/engine.py tests/unit/runtime/pipe/test_pipe.py
• Two-process CPU/Gloo numerical comparison for gradient_accumulation_steps=1,2,8

[tohtana]

Hi @Functionhx,
Thank you for submitting this PR! This looks good to me.
I know this type of bug is very hard to find and fix. I appreciate your continuous contribution.

[PKUWZP]

LGTM, thanks for the PR!