Track token scores (#571)

* add export of per-token scores (ys, lm, context)

- for best path of the modified-beam-search decoding of transducer

* refactoring JSON export of OnlineRecognitionResult, extending pybind11 API of OnlineRecognitionResult

* export per-token scores also for greedy-search (online-transducer)

- export un-scaled lm_probs (modified-beam search, online-transducer)
- polishing

* fill lm_probs/context_scores only if LM/ContextGraph is present (make Result smaller)

sherpa-onnx/csrc/online-transducer-decoder.h

@@ -26,6 +26,10 @@ struct OnlineTransducerDecoderResult {
   /// timestamps[i] contains the output frame index where tokens[i] is decoded.
   std::vector<int32_t> timestamps;
 
+  std::vector<float> ys_probs;
+  std::vector<float> lm_probs;
+  std::vector<float> context_scores;
+
   // Cache decoder_out for endpointing
   Ort::Value decoder_out;