Heap OOB read in `FusedBatchNorm` kernels

TensorFlow is an open source platform for machine learning. In affected versions the implementation of `FusedBatchNorm` kernels is vulnerable to a heap OOB access. The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, as these are also affected and still in supported range.