feat: add layerwise split kv cache sharding.

[shifengmin]

Summary

• Add --layerwise_split_size to shard persistent KV by a static layer owner inside each attention TP group, while non-owners share one scratch layer so PD-Disaggregation stay compatible.
• Depends on the companion ATB PR that implements owner attention.

Dependency

Depends on xLLM-AI/xllm-atb-layers#14

Related Issues

Change Type

• Bug fix
• New feature
• Performance improvement
• Refactor
• Documentation
• Test
• Build or CI

Pull Request Checklist

PR Title and Commit Messages

• The PR title and each commit message follow the xLLM commit format: <type>: <subject>.

Pre-commit Checks

• I have installed pre-commit by running pip install pre-commit or an equivalent command.
• I have installed the hooks with pre-commit install.
• I have run pre-commit run --all-files and fixed any reported issues.

Changed-file validation passed:

pre-commit run --files tests/core/runtime/acl_graph_executor_test.cpp \
  xllm/core/runtime/acl_graph_executor_impl.cpp \
  xllm/core/runtime/acl_graph_executor_impl.h
clang-format.............................................................Passed

Self Review

• I have self-reviewed the code according to .agents/skills/code-review/references/custom-code-style.md, especially code written or assisted by AI.
• I have rebased this PR onto the latest main branch.

Build and Test Coverage

• Tests have been added or updated as needed.
• CUDA: python setup.py build test has passed on a CUDA machine.
• NPU: a full python setup.py build --device npu completed successfully on an Ascend development machine.
• MLU: python setup.py build test has passed on an MLU machine.

Reviewer Notes

This intentionally falls back to eager execution instead of caching one graph per DP distribution. The latter can create an unbounded graph cache for normal scheduling variance.