Hybrid Engine Refactor and Llama Inference Support

[cmikeh2]

This PR introduces a number of features and bugfixes:

• The Hybrid Engine integration with Containers has been refactored. Models that support the Hybrid Engine now inherit from a feature container, either the HybridEngineContainer itself or something more specialized for the particular model architecture.
• Llama support for both inference and RLHF training acceleration with Hybrid Engine support
• Additional BF16 compilation support
• Additional unit test coverage for new operators and data types
• Clean up of unused code