Generate Image that summarizes the convo + create an image for agent based off of the description

[mcp97]

[mcp97]

Currently it is in a WIP branch. Currently it generates the image but it isn't the best image so we need to iterate on the branch . We need to add logic to get blob from url and download it and store it properly: #53

[mcp97]

@0x369D could potentially finish this through as I have hopped onto the architecture refactor