phsl / new-api

根本原因：
1. 在OaiStreamHandler流式处理函数中，调用applyUsagePostProcessing(info, usage, nil)时传入的responseBody为nil，导致无法从响应体中提取缓存tokens。
2. 两个渠道的cached_tokens位置不同：
  - 智普：标准位置 usage.prompt_tokens_details.cached_tokens
  - Moonshot：非标准位置 choices[].usage.cached_tokens

处理方案：
1. 传递body信息到applyUsagePostProcessing中
2. 拆分智普和Moonshot的解析，并为Moonshot单独写一个解析方法。

- wan2.6系列生图与编辑，适配多图生成计费
- wan2.5系列生图与编辑
- z-image-turbo生图，适配prompt_extend计费

* wip ionet integrate

* wip ionet integrate

* wip ionet integrate

* ollama wip

* wip

* feat: ionet integration & ollama manage

* fix merge conflict

* wip

* fix: test conn cors

* wip

* fix ionet

* fix ionet

* wip

* fix model select

* refactor: Remove `pkg/ionet` test files and update related Go source and web UI model deployment components.

* feat: Enhance model deployment UI with styling improvements, updated text, and a new description component.

* Revert "feat: Enhance model deployment UI with styling improvements, updated text, and a new description component."

This reverts commit 8b75cb5bf0d1a534b339df8c033be9a6c7df7964.

* fix: Refine openrouter error

* fix: Refine openrouter error

* fix: openrouter test max_output_token

* fix: optimize messages

* fix: maxToken unified to 16

* fix: codex系列模型使用 responses接口

* fix: codex系列模型使用 responses接口

* fix: 状态码非200打印错误信息

* fix: 日志里没有报错的响应体

fix(gemini): handle minimal reasoning effort budget

fix: 支持传入system_instruction和systemInstruction两种风格系统提示词参数名

fix: add warning for pass through body

fix: 支持小写bearer和Bearer后带多个空格 && 修复 WSS预扣费错误提取key的问题

i18next uses ':' as namespace separator by default, causing URLs like
'https://api.openai.com' to be incorrectly parsed as namespace 'https'
with key '//api.openai.com', resulting in truncated display.

Setting nsSeparator to false fixes this issue since the project doesn't
use multiple namespaces.

- Only short-circuit when there are no missing models AND no overwrite fields requested
- Preserve overwrite behavior even when the missing-model list is empty
- Always return empty arrays (not null) for list fields to keep API responses stable
- Clarify SyncUpstreamModels behavior in comments (create missing models + optional overwrite updates)

fix: ensure overwrite works correctly when no missing models

feat: pyroscope integrate

fix: revert model ratio

fix: 修复 Anthropic 渠道缓存计费错误

add stricter validation for content_block_start corresponding to
tool call
and fix the crash issue when Claude Code is processing tool call

修复claudeResponse流式请求空指针Panic

为Moonshot添加缓存tokens读取逻辑。其与智普V4的逻辑相同，所以共用逻辑

🔗 docs(readme): update documentation links to new site routing

Keep new-site links (/{lang}/docs/...) where matching pages exist in the current docs repo
Revert links that have no equivalent in the new docs to the legacy paths on doc.newapi.pro:
Google Gemini Chat
Midjourney-Proxy image docs
Suno music docs
Apply the same rule consistently across all README translations (zh/en/ja/fr)

- Replace legacy `docs.newapi.pro` paths with the new `/{lang}/docs/...` structure across all README translations
- Point key sections (installation, env vars, API, support, features) to their new locations
- Ensure language-specific links use the correct locale prefix (zh/en/ja) and keep FR aligned with English routes

fix: 模型设置增加针对Vertex渠道过滤content[].part[].functionResponse.id的选项，默认启用