[data] Fix minicpmv/o dpo training (#6657)

* fix template name

* tiny fix

* support minicpm-o-2.6

* support inference of minicpmv

* update readme

* support dpo of minicpmv

src/llamafactory/data/collator.py

@@ -155,7 +155,7 @@ def __call__(self, features: Sequence[Dict[str, Any]]) -> Dict[str, "torch.Tenso
         if "image_bound" in features:  # for minicpmv inputs
             bsz, seq_length = features["input_ids"].shape
             features["position_ids"] = torch.arange(seq_length).long().repeat(bsz, 1)
-            return {"data": features, "labels": features["labels"]}
+            return {"data": features, "input_ids": features["input_ids"], "labels": features["labels"]}
 
         return features