Fix offloading saved tensor interaction with checkpointing

[janeyx99]

Context

What is the purpose of this PR? Is it to

• add a new feature
• fix a bug
• update tests and/or documentation
• other (please add here)

Please link to any issues this PR addresses.
Hopefully fixes #1867 once and for all

Needs pytorch/pytorch#139109 to land!

Changelog

What are the changes made in this PR?

• Fixes checkpoint interaction with offloading. Prior, if an unpacked tensor is cached by activations checkpointing as a saved variable, it may be freed too early. This PR uses hooks (not currently in core yet, so we'll have to wait for that) to be able to coordinate this save properly.
• I should also add a test case, but that'll take me a while to contrive lol.

Test plan

Please make sure to do each of the following if applicable to your PR. If you're unsure about any one of these just ask and we will happily help. We also have a contributing page for some guidance on contributing.

• run pre-commit hooks and linters (make sure you've first installed via pre-commit install)
• add unit tests for any new functionality
• update docstrings for any new or updated methods or classes
• run unit tests via pytest tests
• run recipe tests via pytest tests -m integration_test
• manually run any new or modified recipes with sufficient proof of correctness
• include relevant commands and any other artifacts in this summary (pastes of loss curves, eval results, etc.)

UX

If your function changed a public API, please add a dummy example of what the user experience will look like when calling it.
Here is a docstring example
and a tutorial example

• I did not change any public API
• I have added an example to docs or docstrings

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchtune/1913

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 2 Cancelled Jobs

As of commit 83e6b51 with merge base 74139c9 ():

NEW FAILURE - The following job has failed:

• GPU tests / gpu_test (3.9, stable) (gh)
  FAILED tests/torchtune/training/test_activation_offloading.py::test_offloading_works_with_view_outputs - AttributeError: module 'torch.utils.checkpoint' has no attribute '_register_checkpoint_saved_tensor_hook'

CANCELLED JOBS - The following jobs were cancelled. Please retry:

• GPU tests / gpu_test (3.10, stable) (gh)
  ##[error]The operation was canceled.
• GPU tests / gpu_test (3.11, stable) (gh)
  torchtune/training/_activation_offloading.py:305: AttributeError

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[janeyx99]

Closing in favor of #1936