feat: Support gemini embeddings (text-embedding-004,embedding-001)

[mxdlzg]

close #1313
close #868

我已确认该 PR 已自测通过，相关截图如下：
（此处放上测试通过的截图，如果不涉及前端改动或从 UI 上无法看出，请放终端启动成功的截图）

[SYS] 2024/05/28 - 14:34:52 | One API v0.0.0 started 
[SYS] 2024/05/28 - 14:34:52 | using MySQL as database 
[SYS] 2024/05/28 - 14:34:52 | database migration started 
[SYS] 2024/05/28 - 14:34:52 | database migrated 
[SYS] 2024/05/28 - 14:34:52 | REDIS_CONN_STRING not set, Redis is not enabled 
[SYS] 2024/05/28 - 14:34:52 | using theme berry 
[SYS] 2024/05/28 - 14:34:52 | initializing token encoders 
[SYS] 2024/05/28 - 14:34:57 | token encoders initialized 
[SYS] 2024/05/28 - 14:35:50 | model ratio not found: text-embedding-004 
[INFO] 2024/05/28 - 14:35:50 | 202405281435508260323702375031 | user 1 has enough quota 499999972164332, trusted and no need to pre-consume 
[GIN] 2024/05/28 - 14:35:50 | 202405281435508260323702375031 | 200 |  426.223764ms | ip |    POST /v1/embeddings
[INFO] 2024/05/28 - 14:35:50 | 202405281435508260323702375031 | record consume log: userId=1, channelId=4, promptTokens=0, completionTokens=0, modelName=text-embedding-004, tokenName=FastChat, quota=0, content=模型倍率 30.00，分组倍率 1.00，补全倍率 1.00 
[SYS] 2024/05/28 - 14:36:07 | model ratio not found: text-embedding-004 
[INFO] 2024/05/28 - 14:36:07 | 2024052814360737548440985902243 | user 1 has enough quota 499999972164332, trusted and no need to pre-consume 
[GIN] 2024/05/28 - 14:36:07 | 2024052814360737548440985902243 | 200 |  353.767544ms | ip |    POST /v1/embeddings
[INFO] 2024/05/28 - 14:36:07 | 2024052814360737548440985902243 | record consume log: userId=1, channelId=4, promptTokens=0, completionTokens=0, modelName=text-embedding-004, tokenName=FastChat, quota=0, content=模型倍率 30.00，分组倍率 1.00，补全倍率 1.00

模型名字在测试时有映射（google-text-embedding-004->text-embedding-004）
多条：

单条：

[songquanpeng]

Thx~

[yoursbest]

@mxdlzg , @songquanpeng 你们好，

感谢提供gemini embedding的adaptor！

我发现一个问题，

• 直接通过curl访问one-api可以完美进行embedding
• 如果通过第三方工具或者langchain包来调用会失败，例如：

from langchain_openai import OpenAIEmbeddings


embeddings_model = OpenAIEmbeddings(
    openai_api_base=openai_api_base,
    openai_api_key=openai_api_key,
    model="text-embedding-004",
)

embeddings = embeddings_model.embed_query('hello')

出现错误 BadRequestError: Error code: 400 - {'error': {'message': '* BatchEmbedContentsRequest.requests: requests must not be empty\n (request id: 2024062618344384827435121138789)', 'type': '', 'param': '', 'code': 400}}

这个问题不知是否可以解决，谢谢！