fix: avoid cleanup errors for partially initialized LlamaModel (#2173)

* Add attribute check for sampler in close method

This solves a bug I uncovered, that causes an AttributeError if constantly re-initializing a model in a loop and Python garbage collects it, such as testing the highest GPU layer count you can go before CUDA OOMs.

* fix: avoid cleanup errors for partial model init

---------

Co-authored-by: abetlen <abetlen@gmail.com>