[Doc] Document checkpointing to resume RL training

[YeonwooSung]

Description

Adds a Resume training guide to docs/source/reference/checkpoint.rst so users can save and restore an RL agent with the existing :class:~torchrl.checkpoint.Checkpoint`` API.

The new section:

• Lists the components a resume checkpoint should contain (training model, optimizer, target networks; collector policy / env-transform state / frame count; replay-buffer writer, sampler, transforms, and storage; RNG) and why each one matters in RL.
• Shows constructing Checkpoint(policy=..., optimizer=..., replay_buffer=..., collector=..., loss_module=..., exploration=..., rng=GlobalRNGState()), then save / load with components={...} and map_location.
• Clarifies that the payload is the TensorDictModule / state_dict (or the buffer's dump / load), not a pickled mystery container.
• Cross-links the shared-module case from [Feature Request] Documentation about weight update propagation #3032: if the loss holds the same module, save the policy or the loss once; do not persist two independent copies of the online weights.
• Contrasts checkpoint (interrupt and resume) with export (deploy for inference) and links the export tutorial.

Includes a compact CartPole DQN snippet a reader can copy.

Motivation and Context

close #3044

The export tutorial is inference-only. Users asked for a copy-paste guide to interrupt training and continue it later. Checkpoint already exists; this PR documents how to use it rather than inventing a parallel API.

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

Types of changes

• Documentation (update in the documentation)

Checklist

• 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/4230

• 📄 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 f9a17e7 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 f64fe16c478f300f86670213e2a05cb9648f0cd8497ac1550dde5caa80853423 /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-cpu (3.14, bulk) / linux-job (gh)
  AssertionError: assert False

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