Support image input in the chat completion request

[Youho99]

Tested with a single image

This pull request responds to issue #54

It allows you to take into account the architecture of the OpenAI API request with an image

Example on the OpenAI documentation:

curl https://api.openai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "model": "gpt-4-turbo",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "text",
            "text": "What'\''s in this image?"
          },
          {
            "type": "image_url",
            "image_url": {
              "url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"
            }
          }
        ]
      }
    ],
    "max_tokens": 300
  }'

The code has not been prettyfied, so we need to review that

[lhenault]

Thanks for your work, will happily review this once you think it's ready (and passing the pre-commit check). If you have a working example for VLM / image processing to share, that would be a nice addition to the existing ones.

[Youho99]

Don't use grpcio and grpcio-tools 1.65.0 version (remised version)

I don't know how to modify it in the poetry requirements

[Youho99]

I just modified the rules regarding the versions of grpcio and grpcio-tools in the toml, and I regenerated the poetry.lock

Since this is my first time doing this, I would like to request special attention on this.

[Youho99]

I will provide an example of using my feature in a second step (in another PR I think)

[Youho99]

@lhenault I think you can review this PR (and change the version accordingly) :)

[lhenault]

Hey @Youho99 !

I tried your changes the other day and encountered a few issues, but probably because of me. Thanks again for your PR and sorry for the delay, it's very much appreciated. 😌

Let me have another look soon (and if you have a working example for image inputs that might speed up things).

[Youho99]

@lhenault

In the next few days I'll get back to it, and provide an example.

Let me know if you have any problems.

[Youho99]

@lhenault Hello and happy new year!

After a fews days (lol), i have finally produce an example for the image support.

Well, this one is not in the format of the examples already present in the library. We can do this work later.

Here is the project:
https://github.com/Youho99/phi-3_5-vision-onnx-simpleai

[Youho99]

@lhenault any update ?

[lhenault]

Hey sorry I somehow missed this and the previous update. I'll have a look at it soon. Thanks a lot for the submission!

[Youho99]

@lhenault
Can you reviex this PR ?