[AZ-180] Refactor inference and engine factory for improved model handling

- Updated the autopilot state to reflect the current task status as in progress.
- Refactored the inference module to streamline model downloading and conversion processes, replacing the download_model method with a more efficient load_source method.
- Introduced asynchronous model building in the inference module to enhance performance during model conversion.
- Enhanced the engine factory to include a new method for building and caching models, improving error handling and logging during the upload process.
- Added calibration cache handling in the Jetson TensorRT engine for better resource management.

Made-with: Cursor

_docs/_autopilot_state.md

@@ -4,8 +4,8 @@
 flow: existing-code
 step: 8
 name: New Task
-status: not_started
-sub_step: 0
+status: in_progress
+sub_step: 1 — Gather Feature Description
 retry_count: 0
 
 ## Cycle Notes