Add routed_scaling_factor to MoE grouped topk (#23123)

Signed-off-by: Xin Yang <xyangx@amazon.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

vllm/model_executor/models/deepseek_v2.py

@@ -160,6 +160,7 @@ class DeepseekV2MoE(nn.Module):
             topk_group=config.topk_group,
             prefix=f"{prefix}.experts",
             scoring_func=config.scoring_func,
+            routed_scaling_factor=self.routed_scaling_factor,
             e_score_correction_bias=self.gate.e_score_correction_bias,
             enable_eplb=self.enable_eplb,
             num_redundant_experts=self.n_redundant_experts)