Character creation support from reference images

Assumes the use of the reference image function. Does not consider the use of blurry images.

Only considers creating one character setting at a time. This is because it is difficult to recognize more than one person at a time due to the use of the reference image function.
It is particularly specialized for the first response and is not good at continuing conversations. It is unlikely to break down until about the second time.
For the same reason, if you want to create settings for different aspects or situations of an already created character, or for related characters, you can also output them in the first response by pasting the completed profile or settings into your user notes and presenting them to the AI.

Confirmed operation on gemini-based models with high image recognition capabilities.