Fixes #158 : avoid duplicate add Qwen3.5 shared expert output

[johnny-zjh]

What does this PR do?

Type of change: Bug fix

Overview: Fixes duplicate shared-expert accumulation in the Qwen3.5 MoE forward path. Qwen3SparseMoeBlock.forward() already combines the routed-expert output with the gated shared-expert output. Qwen3_5SparseMoeBlock.forward() previously called the base implementation and then added the same shared-expert output again, casue erros;

Usage

No user-facing API or workflow changes are required. Export Qwen3.5 MoE models using the existing documented workflow.

🚀 Pull Request Checklist

Thank you for contributing to TensorRT Edge-LLM! Before we review your pull request, please make sure the following items are complete.
Please also refer to Contributor guidelines for general guidelines.

✅ Pre-commit Checks

• ✅ I have installed pre-commit by running pip install pre-commit.
• ✅ I have installed the hooks with pre-commit install.
• ✅ I have run the hooks manually with pre-commit run --all-files and fixed any reported issues.

🧪 Tests

• Tests have been added or updated as needed.
• All tests are passing.

📄 Documentation

• Documentation updates are not required because this change only corrects internal model execution behavior.

⚙️ Compatibility

• ✅ The change is backward compatible