Add AVX_VNNI support for intel x86 processors

[freebie1101]

Prerequisites

Please answer the following questions for yourself before submitting an issue.

• I am running the latest code. Development is very rapid so there are no tagged versions as of now.
• I carefully followed the README.md.
• [x ] I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new bug or useful enhancement to share.

Feature Description

Enable AVX_VNNI ( we already have AVX512 ). Enable support for AVX_VNNI will make intel processor able to use int8 computation.

Motivation

Intel x86 processors are still very popular among consumer and data centers hardwares. Enable AVX_VNNI will make the performance of llama cpp on these processors better.

[riverzhou]

Same question.

[sorasoras]

I build windows rocm build of llamacpp, it does support VNNI.

[tikikun]

I build windows rocm build of llamacpp, it does support VNNI.

why rocm support AVX_VNNI ??

[ernesst]

This should be helpful I guess : https://www.intel.com/content/www/us/en/content-details/791610/optimizing-and-running-llama2-on-intel-cpu.html

[tikikun]

^ Based on this comment i have implemented this PR
#4589

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.