Add flag to apply DeduplicateHashedInitializersPass post graph surgery

[qti-ashimaj]

Describe your changes

Add flag to apply DeduplicateHashedInitializersPass post graph surgery.
With the DeduplicateHashedInitializersPass, the VRAM usage for onnx static quantization increased multifold, hence adding an option to keep this pass.
For Qwen2.5-1.5B-Instruct model, using DeduplicateHashedInitializersPass needs ~58GB VRAM while without DeduplicateHashedInitializersPass needs only ~14GB VRAM for static quantization

Checklist before requesting a review

• Add unit tests for this change.
• Make sure all tests can pass.
• Update documents if necessary.
• Lint and apply fixes to your code by running lintrunner -a
• Is this a user-facing change? If yes, give a description of this change to be included in the release notes.

(Optional) Issue link