Add support for INT8 quantization in model export via engine

[Surya-Rayala]

Search before asking

• I have searched the BoxMOT issues and found no similar enhancement requests.

Description

boxmot/appearance/reid/export.py

Can we add an argument to enable INT8 quantization when exporting the model with the engine?

Use case

No response

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[mikel-brostrom]

I do not have time to look into this at the moment but feel free leave a PR ☺️