[megatron] fix: mbridge load optimizer dist_ckpt (verl-project#3850)

### What does this PR do?

fix training resume from mbridge load optimizer dist_ckpt.

This PR fix verl-project#3517


> Add **concise** overview of what this PR aims to achieve or
accomplish. Reference related GitHub issues and PRs that help with the
review.

### Checklist Before Starting

- [ ] Search for similar PRs. Paste at least one query link here: ...
- [ ] Format the PR title as `[{modules}] {type}: {description}` (This
will be checked by the CI)
- `{modules}` include `fsdp`, `megatron`, `sglang`, `vllm`, `rollout`,
`trainer`, `ci`, `training_utils`, `recipe`, `hardware`, `deployment`,
`ray`, `worker`, `single_controller`, `misc`, `perf`, `model`, `algo`,
`env`, `tool`, `ckpt`, `doc`, `data`
- If this PR involves multiple modules, separate them with `,` like
`[megatron, fsdp, doc]`
  - `{type}` is in `feat`, `fix`, `refactor`, `chore`, `test`
- If this PR breaks any API (CLI arguments, config, function signature,
etc.), add `[BREAKING]` to the beginning of the title.
  - Example: `[BREAKING][fsdp, megatron] feat: dynamic batching`

### Test

> For changes that can not be tested by CI (e.g., algorithm
implementation, new model support), validate by experiment(s) and show
results like training curve plots, evaluation results, etc.

### API and Usage Example

> Demonstrate how the API changes if any, and provide usage example(s)
if possible.

```python
# Add code snippet or script demonstrating how to use this
```

### Design & Code Changes

> Demonstrate the high-level design if this PR is complex, and list the
specific changes.

### Checklist Before Submitting

> [!IMPORTANT]
> Please check all the following items before requesting a review,
otherwise the reviewer might deprioritize this PR for review.

- [ ] Read the [Contribute
Guide](https://github.com/volcengine/verl/blob/main/CONTRIBUTING.md).
- [ ] Apply [pre-commit
checks](https://github.com/volcengine/verl/blob/main/CONTRIBUTING.md#code-linting-and-formatting):
`pre-commit install && pre-commit run --all-files --show-diff-on-failure
--color=always`
- [ ] Add / Update [the
documentation](https://github.com/volcengine/verl/tree/main/docs).
- [ ] Add unit or end-to-end test(s) to [the CI
workflow](https://github.com/volcengine/verl/tree/main/.github/workflows)
to cover all the code. If not feasible, explain why: ...
- [ ] Once your PR is ready for CI, send a message in [the `ci-request`
channel](https://verl-project.slack.com/archives/C091TCESWB1) in [the
`verl` Slack
workspace](https://join.slack.com/t/verl-project/shared_invite/zt-3855yhg8g-CTkqXu~hKojPCmo7k_yXTQ).
(If not accessible, please try [the Feishu group
(飞书群)](https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=772jd4f1-cd91-441e-a820-498c6614126a).)

---------

Co-authored-by: Gao Shiyuan <gaoshiyuan@baidu.com>

Author: ccilery

verl/utils/checkpoint/megatron_checkpoint_manager.py

@@ -231,7 +231,11 @@ def get_checkpoint_name(
         return os.path.join(common_path, basename)
 
     def generate_state_dict(
-        self, generate_model: bool = True, generate_optimizer: bool = True, generate_extra: bool = True
+        self,
+        generate_model: bool = True,
+        generate_optimizer: bool = True,
+        generate_extra: bool = True,
+        is_loading: bool = False,
     ):
         # For save dist checkpointing
         state_dict = {}
@@ -252,7 +256,7 @@ def generate_state_dict(
         # Optimizer State Dict
         if generate_optimizer:
             torch.distributed.barrier()
-            optimizer_sharded_states = self.optimizer.sharded_state_dict(state_dict)
+            optimizer_sharded_states = self.optimizer.sharded_state_dict(state_dict, is_loading=is_loading)
             state_dict["optimizer"] = optimizer_sharded_states
 
             if self.lr_scheduler is not None:
@@ -292,11 +296,20 @@ def load_checkpoint(self, local_path: str, hdfs_path: str = None, del_local_afte
         if local_path is not None:
             assert os.path.exists(local_path), f"Checkpoint path {local_path} does not exist."
 
+        # For load optimizer dist_ckpt
+        import transformer_engine
+
+        torch.serialization.add_safe_globals([torch.optim.AdamW])
+        torch.serialization.add_safe_globals([transformer_engine.pytorch.optimizers.fused_adam.FusedAdam])
+
         dist_checkpoint_path = get_dist_checkpoint_path(local_path)
 
         # Get State Dict for loading
         sharded_state_dict = self.generate_state_dict(
-            self.should_load_model and self.use_dist_checkpointing, self.should_load_optimizer, self.should_load_extra
+            self.should_load_model and self.use_dist_checkpointing,
+            self.should_load_optimizer,
+            self.should_load_extra,
+            is_loading=True,
         )
         log_with_rank(f"Generated state dict for loading: {sharded_state_dict.keys()}", rank=self.rank, logger=logger)