feat: add generationConfig of Google Gemini & fix: Gemini abort bug

[LyuLumos]

Description

• feat: add support for Google Gemini configuration: temperature maxOutputTokens topP topK
• fix: the abort button does not work when using Gemini.

Linked Issues

None.

Additional context

There are two similar parameters in the Gemini settings, the first is maxTokens. This parameter serves the same purpose as before, is used to cut the history length.

The second one is maxOutputTokens, which is an offical parameter for Gemini. In the Google docs, Input token limit is a fixed value, (e.g. 30720 for gemini-pro). The user can only change maxOutputTokens.

• Link: Gemini Doc

[netlify[bot]]

❌ Deploy Preview for chatgpt-demo-v2 failed.

Name	Link
🔨 Latest commit	aefed85526a2947e143351eeee7c11f958c10329
🔍 Latest deploy log	https://app.netlify.com/sites/chatgpt-demo-v2/deploys/65b77fe65cc3250008b35bbc

[ddiu8081]

Thanks!