[Bugfix] Merge MM embeddings by index instead of token IDs (#16229)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Roger Wang <hey@rogerw.io>

vllm/config/model.py

@@ -509,9 +509,14 @@ class ModelConfig:
                 else:  # task == "auto"
                     pass
             else:
+                debug_info = {
+                    "architectures": architectures,
+                    "is_generative_model": is_generative_model,
+                    "is_pooling_model": is_pooling_model,
+                }
                 raise AssertionError("The model should be a generative or "
                                      "pooling model when task is set to "
-                                     f"{self.task!r}.")
+                                     f"{self.task!r}. Found: {debug_info}")
 
             self.runner = runner
             self.convert = convert