[Fix] Support fp32 param preservation during FSDP save and load

[HAOCHENYE]

Stack from ghstack (oldest at bottom):

• [Feature] Add SP support for roll_sequence_context in MTP #1629
• fixup! [Feature] Add Multi-Token Prediction (MTP) module implementation #1628
• fixup! [Feature] Add Multi-Token Prediction (MTP) module implementation #1627
• [Refactor] Rename CELossContext to LMHeadLossContext and refactor loss context base class #1571
• [Feature] Add Multi-Token Prediction (MTP) module implementation #1570
• [Refactor] Refactor loss context API to support multiple loss types #1569
• -> [Fix] Support fp32 param preservation during FSDP save and load #1626

---

• Add fp32_keys_pattern to HFSaveCfg to specify params that should
  be saved in fp32 regardless of the global save dtype
• Add _fully_shard() to BaseModel which wraps fully_shard() and
  distributes matched params as Replicate DTensors so FSDP ignores them
• Replace all direct fully_shard() call sites with self._fully_shard()
• Add _get_save_dtype() to select fp32 per-param at save time
• Fix load path in _load_same_hf_param to skip shard-offset logic for
  Replicate DTensors (only apply for Shard-placed params)
• Fix world_mesh property setter bug: use _world_mesh directly
  instead of assigning through the read-only property
• Fix gradient reduce in MoE.scale_and_reduce_grad to correctly
  identify the Replicate mesh dimension via DTensor placements
• Preserve requires_grad when distributing params in EP and fp32 paths
• Configure Qwen3_5_VLTextMoEConfig with fp32 patterns for
  linear_attn.norm.weight and linear_attn.A_log
• Add test_save_hf_with_mtp to verify round-trip weight preservation