Implement AI TOPS Calculation

[tembo]

Summary

Adds AI compute power (TOPS) calculation for devices by:

• Creating hardware_specs.rs with comprehensive TOPS lookup for CPUs and GPUs
• Adding ai_tops field to Device struct
• Implementing total AI compute calculation in library statistics

Key Changes

• Added detailed hardware AI capabilities lookup for:
  • Apple Silicon Neural Engine
  • NVIDIA Tensor Core GPUs
  • AMD Matrix Engine GPUs
  • Intel NPUs
• Integrated TOPS calculation in device detection and library statistics
• Added comprehensive test coverage for different hardware scenarios

Technical Details

• Supports multiple precision modes (INT8, FP16)
• Handles single and multi-GPU configurations
• Case-insensitive hardware model matching
• Gracefully handles unknown hardware

Considerations

• Lookup table can be expanded over time
• Provides foundation for future AI compute tracking

Want tembo to make any changes? Add a review or comment with @tembo and i'll get back to work!

[tembo]

Requesting review from @jamiepine who has experience with the following files modified in this PR:

• core/src/domain/mod.rs
• core/src/library/mod.rs
• core/src/domain/device.rs
• core/src/library/config.rs