Remove RNNTL unused vars (#2142)

Summary:
remove unnecessary RNNT Loss variables and comment as indicated in https://github.com/pytorch/audio/issues/1479 review comments
(will follow up on `workspace` comments separately depending on complexity)

Pull Request resolved: https://github.com/pytorch/audio/pull/2142

Reviewed By: mthrok

Differential Revision: D33433764

Pulled By: carolineechen

fbshipit-source-id: be0ecb77dabd63d733f0d33ff258eae32305eeaf

torchaudio/csrc/rnnt/cpu/cpu_kernels.h

@@ -444,7 +444,6 @@ void ComputeAlphas(
         TensorView<CAST_DTYPE>({maxT, maxU}, alphas + b * maxT * maxU));
   }
 
-  std::vector<CAST_DTYPE> scores(B << 1);
   //#pragma omp parallel for
   for (int i = 0; i < B; ++i) { // use max 2 * B threads.
     ComputeAlphaOneSequence<DTYPE>(
@@ -481,9 +480,8 @@ void ComputeBetas(
         TensorView<CAST_DTYPE>({maxT, maxU}, betas + b * maxT * maxU));
   }
 
-  std::vector<CAST_DTYPE> scores(B << 1);
   //#pragma omp parallel for
-  for (int i = 0; i < B; ++i) { // use max 2 * B threads.
+  for (int i = 0; i < B; ++i) {
     ComputeBetaOneSequence<DTYPE>(
         options,
         /*logProbs=*/seqlogProbs[i],