Optimize Platform + Restructure

[AndreSlavescu]

• [x] Restructure Makefile (automate detection of compute capability)
• [ ] Optimize existing kernels

[jvhs0706]

Hi Andre, your optimizations seem useful! Do you want us to merge now, or do you want to do that after you further optimize existing kernels?

[AndreSlavescu]

Hi Andre, your optimizations seem useful! Do you want us to merge now, or do you want to do that after you further optimize existing kernels?

Hi Haochen,

I will be working on optimizing the kernels as well. For now the changes are for automatically detecting compute capability which eliminates the need for filling that in manually.

I am also trying to discover what CUDA versions work with this setup, because I was having problems with 11.7+

For now, I will just leave this as a draft PR and promptly make the updates to speedup the kernels.

[hongyanz]

@AndreSlavescu Hi Andre, can you contact me by my email?