add A2 cpu inference

[MarcelWilnicki]

Hi,

A proposal of A2 blueprint. The blueprint is similar to this blueprint: https://github.com/oracle-quickstart/oci-ai-blueprints/blob/main/docs/sample_blueprints/model_serving/cpu-inference/cpu-inference-mistral-vm.json

and aimed at reaching similar performance for mistral:7b-instruct-q8_0 and llama3.1:8b-instruct-q8_0 models.

from our internal benchmarks we got such numbers for A2:

mistral:7b-instruct-q8_0
b=1, t=8, tg_throughput=6.27, pp_throughput=446.95, ttft=0.15

llama3.1:8b-instruct-q8_0
b=1, t=8, tg_throughput=5.63, pp_throughput=419.10, ttft=0.20

and such numbers for E4:

mistral:7b-instruct-q8_0
b=1, t=4, tg_throughput=6.91, pp_throughput=470.62, ttft=0.15

llama3.1:8b-instruct-q8_0
b=1, t=4, tg_throughput=6.54, pp_throughput=427.20, ttft=0.17

The main difference in the blueprint is 8 cores for A2 vs 4 cores for E4.

[oracle-contributor-agreement]

Thank you for your pull request and welcome to our community! To contribute, please sign the Oracle Contributor Agreement (OCA).
The following contributors of this PR have not signed the OCA:

• PR author: MarcelWilnicki
• marcel.wilnicki@gmail.com (@MarcelWilnicki)

To sign the OCA, please create an Oracle account and sign the OCA in Oracle's Contributor Agreement Application.

When signing the OCA, please provide your GitHub username. After signing the OCA and getting an OCA approval from Oracle, this PR will be automatically updated.

If you are an Oracle employee, please make sure that you are a member of the main Oracle GitHub organization, and your membership in this organization is public.