[NPU] Fix Ascend JSD UT failures and register LayerNorm dispatch

[sunyi0505]

Summary

Unit tests for JSD / FusedLinearJSD on Ascend failed, mainly due to signature misalignment with the NVIDIA frontend, and invalid kernel launches when all samples are ignored. The ascend-triton dispatcher registration for LayerNorm is also added so that the NPU can automatically select the UB-aware implementation.

JSD / FusedLinearJSD signature alignment: NVIDIA's LigerJSDFunction / LigerFusedLinearJSDFunction accepts jsd_impl / jsd_mode (apply(..., None, None) is used in tests). The Ascend kernel is self-contained and does not use the inner dispatcher, so it simply accepts and ignores these two parameters, and the number of backward return values is supplemented accordingly; supports_inner_impl_dispatch is set to False for JSD.

Short-circuit for all-ignored case: When n_non_ignore == 0, skip kernel launch and directly return zero loss and zero gradients. This covers cases such as test_correctness_all_ignored to prevent division-by-zero / invalid grid.

LayerNorm registration: Add ops/backends/_ascend/layer_norm.py and register it in declare_op_locations("layer_norm") within functional.py. Otherwise, auto-selection only detects the NVIDIA-triton implementation for CUDA and cannot use _ascend.ops.layer_norm.

Testing Done

pytest tests/transformers

• Hardware Type: Atlas 800T A3(X86)
• run make test to ensure correctness
• run make checkstyle to ensure code style
• run make test-convergence to ensure convergence

[sunyi0505]

@Tcc0403 This pr is ready for review. Could you please review this PR? Thanks!

[Tcc0403]

Just saw an update, is ci good now?

[sunyi0505]

Just saw an update, is ci good now?

The update only rebases the code from the main branch, with no other changes. CI passes.

[Tcc0403]

LGTM