OpenNMT-tf 2.8.1

Fixes and improvements

• Disable dropout in layers that are frozen by the freeze_layers parameter
• Fix batch size autotuning that ignored the mixed precision flag
• Fix sparse gradients that were unnecessarily converted to dense gradients in mixed precision training
• Only compute the gradients global norm every save_summary_steps steps to save a few computation during training
• Simplify some ops in the inference graph