[Doc] Document LossModule and collector weight sharing

[YeonwooSung]

Description

Document how LossModule and collectors share policy weights, and when collector.update_policy_weights_() is actually required.

The objectives landing page (and the DDPG tutorial) said losses "contain a copy of the trainable parameters". That wording made it look like an optimizer step on the loss would not update the original module, and that a collector always needed an explicit sync. Maintainer answers on #3032 already settled the semantics; this PR writes them down.

• Passing a module into a LossModule does not copy it. The same parameters are used in-place; optimizer steps on the loss update the original module.
• A CPU policy passed to a collector is moved to shared memory in-place. A CUDA policy is already shared.
• update_policy_weights_() is required when policy_device or device remaps the policy (a .to(...) copy is made). Calling it anyway is good practice but not load-bearing unless devices were remapped.

Cross-links: LossModule / objectives landing page ↔ collectors weight-sync page.

Motivation and Context

Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
You can use the syntax close #15213 if this solves the issue #15213

• I have raised an issue to propose this change (required for new features and bug fixes)

close #3032

Types of changes

What types of changes does your code introduce? Remove all that do not apply:

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds core functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation (update in the documentation)
• Example (update in the folder of examples)

Checklist

Go over all the following points, and put an x in all the boxes that apply.
If you are unsure about any of these, don't hesitate to ask. We are here to help!

• I have read the CONTRIBUTION guide (required)
• My change requires a change to the documentation.
• I have updated the tests accordingly (required for a bug fix or a new feature).
• I have updated the documentation accordingly.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/4231

• 📄 Preview Python docs built from this PR

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

❌ 3 New Failures

As of commit dac9946 with merge base 1d3de3d ():

NEW FAILURES - The following jobs have failed:

• Libs Tests on Linux / unittests-gym (3.10, 12.8) / linux-job (gh)
  RuntimeError: Command docker exec -t 91c5d87b6d9227f0e77f3648e246dafc37cc9928ad5dbf8505b01230c94f0b00 /exec failed with exit code 1
• Unit-tests on Linux / tests-cpu (3.10, collectors) / linux-job (gh)
  test/test_collectors.py::TestCollectorGeneric::test_update_weights[MultiProcessWeightSyncScheme-False-False]
• Unit-tests on Linux / tests-gpu (3.12, 13.0, 3) / linux-job (gh)
  AssertionError: Tensor-likes are not close!

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