Migrate INCConfig for HPU

[yiliu30]

As part of vllm-project/vllm#31716, the INCConfig in vllm-project/vllm will be used for GPU/CPU. However, we still need a placeholder for the INC path in the plugin.

cc @hshen14 @thuang6 @kzawora-intel @xuechendi

[Copilot]

Pull request overview

This PR migrates the HPU-specific INCConfig implementation to maintain compatibility as the main vLLM repository moves INC configuration for GPU/CPU use. A placeholder _FakeINCConfig is introduced to handle the "inc" quantization method in the HPU plugin while delegating actual quantization to unquantized methods.

Key Changes:

• Introduces a new _FakeINCConfig class that acts as a stub for Intel Neural Compressor quantization
• Monkey-patches vLLM's get_quantization_config function to intercept "inc" requests and return the fake config

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
vllm_gaudi/extension/quant.py	New file defining _FakeINCConfig stub that returns unquantized methods for linear and MoE layers
vllm_gaudi/extension/ops.py	Adds monkey-patch function oot_get_quantization_config to override vLLM's quantization config retrieval for "inc" method

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

✅ CI Passed

All checks passed successfully against the following vllm commit:
b3a2bdf1ac90748d58bf8c05f8d0095ede5c7eca

[github-actions]

✅ CI Passed

All checks passed successfully against the following vllm commit:
0d7667419f738e44ce7f9bf311987e15b01970b0