[EHZ] Adding and saving context when translating the same game.

[Erisae]

Is your feature request related to a problem? Please describe.
Although it is game translation, there are no predefined game character images/stances/plot/settings to adjust the translated content. I hope to incorporate and accumulate context during the translation process using OpenAI.

Describe the solution you'd like
First, use a template to ask ChatGPT for specific task/story settings and accumulate the content of that conversation along with subsequent dialogues into a prompt.

Describe alternatives you've considered

At the beginning of the translation, users can optionally provide additional information or current image information to perform multimodal feature fusion and obtain the response from ChatGPT.

Additional context
For example extract web page txt (like wiki) as context input.