feat(recipe): add qwen3-vl-8b NPU recipe with DAPO

[yidingdou]

Title: Adds Qwen3-VL-8B vision-language reasoning NPU recipe with DAPO on GEO3K

Why this recipe?

verl-recipe currently lacks vision-language (VL) reasoning recipes on NPU, despite the growing importance of multimodal reasoning.

This recipe is designed as VL reasoning with verifiable rewards:

Shows how run Qwen3-VL-8B dapo on NPU

Recipe Design

Model: Qwen3-VL-8B-Instruct (full-weight, NPU A2 * 2 nnodes)
Dataset: geo3k

Results

100 steps of DAPO on 2 A2(910B3) nnodes 64GB (~53 h):

算法: DAPO
Baseline (mean@1): 0.438
Test acc best (mean@1): 0.715

[gemini-code-assist]

Code Review

This pull request introduces a new shell script run_dapo_qwen3_vl_8b_fsdp2_npu.sh to configure and execute DAPO training for the Qwen3-VL-8B model on NPU hardware. The script defines various hyperparameters, environment variables, and model configurations. Feedback identifies critical syntax errors where spaces were incorrectly inserted into configuration keys (e.g., actor rollout_ref instead of actor_rollout_ref) and a minor formatting inconsistency regarding line continuation characters.