Refactor OV weight compression call inside from_pretrained

[nikita-savelyevv]

What does this PR do?

Address https://github.com/huggingface/optimum-intel/pull/618#discussion_r1533615673

Changes:

• Make optional save_directory parameter for OVQuantizer.quantize() method
• Move calibration dataset assembly logic from OVModelForCausalLM to OVQuantizer

In a similar fashion in the future I plan to move SD dataset collection logic to OVQuantizer

Before submitting

• [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• [ ] Did you make sure to update the documentation with your changes?
• [ ] Did you write any new necessary tests?

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[nikita-savelyevv]

@AlexKoff88 @echarlaix could you please take a look?