[BUGFix] Resgiter missing FP8 LDMATRIX Instructions for dynamic share…

…d memory (#162)

* Merge branch 'main' of https://github.com/microsoft/BitBLAS into main

* remove debug print

* Refactor Matmul class for improved readability and maintainability

* Refactor Matmul class for improved readability and maintainability

* revert set device

* lint fix

* register fp8 for dynamic