[PyTorch] Refactor activation offloading of quantized tensors.

[pggPL]

Description

The code of activation offloading has some complex logic handling offloading Float8Tensor object - it disassembles the object into the data tensors, then offloads them separately and then assembles them.

I add empty_like(..., device=..., pin_memory=...) method to Float8Tensor, which allow easy CPU backup tensor allocation. This allows me to make the code of the offloading much simpler.

Type of change

• Documentation change (change only to the documentation, either a fix or a new content)
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Infra/Build change
• Code refactoring

Changes

Please list the changes introduced in this PR:

• Change A
• Change B

Checklist:

• I have read and followed the contributing guidelines
• The functionality is complete
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

[pggPL]

/te-ci pytorch

[pggPL]

/te-ci pytorch

[pggPL]

/te-ci pytorch

[pggPL]

/te-ci pytorch

[pggPL]

/te-ci pytorch

[pggPL]

/te-ci pytorch