Bump oneapi-basekit, optimum and openvino

[fakezeta]

Description

Bump to the latest release: openvino 2024.1.0

Selected from release notes:

• Mixtral and URLNet models optimized for performance improvements on Intel® Xeon® processors.
• The preview NPU plugin for Intel® Core™ Ultra processors is now available in the OpenVINO open-source GitHub repository, in addition to the main OpenVINO package on PyPI.
• Significant memory reduction for select smaller GenAI models on Intel® Core™ Ultra processors with integrated GPU.
• (CPU) LLM compilation time and memory footprint have been improved through additional optimizations with compressed embeddings.
• (CPU) Performance of MoE (e.g. Mixtral), Gemma, and GPT-J has been improved further.
• (GPU) LLM first token latency has been improved on both integrated and discrete GPU platforms.
• NPU Device Plugin

Planning to enable NPU selection on transformer backend soon©️

Note for reviewers
It's my first PR bumping all this dependencies please review carefully 😄

• Yes, I signed my commits.

[netlify]

✅ Deploy Preview for localai canceled.

Name	Link
🔨 Latest commit	2a2b15b
🔍 Latest deploy log	https://app.netlify.com/sites/localai/deploys/662ae65bebc1610008336bc3

[fakezeta]

I've added also a change to PERFORMANCE HINT to CUMULATIVE THROUGHPUT improving inference speed noticeably.

[mudler]

I'll have to give a test at this, but I don't see anything wrong in bumping the image

[fakezeta]

Very curious to know how it works on Arc.
Still some issue to sort with Llama3 while Phi3-128k is working like a charm.