[Feature] Support return logprob of generated tokens (#2784)

* online chat support logprobs

* check xpu

* check vl_gpu_model_runner

* only cuda support logprob

* get_worker() check platform

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>

fastdeploy/input/text_processor.py

@@ -309,6 +309,10 @@ class DataProcessor(BaseDataProcessor):
         data_processor_logger.info(f"Processed request {request}")
         return request
 
+    def process_logprob_response(self, token_ids, **kwargs):
+        full_text = self.tokenizer.decode(token_ids, **kwargs)
+        return full_text
+
     def process_response(self, response_dict, **kwargs):
         """
         Preprocess the response