feat(provider): 支持智谱 GLM Coding Plan

[RhoninSeiei]

现有智谱 Provider 使用通用 API。GLM Coding Plan 的套餐额度只通过专用 Coding API 提供，手工替换地址后还会遇到模型列表接口不可用、推理参数与模型版本不匹配等问题。

关联 #7109。

Modifications / 改动点

• 新增独立的 zhipu_coding_plan_chat_completion Provider，保留现有智谱通用 API Provider。
• 默认使用国内 Coding API，同时允许配置全球 Coding API。
• 使用固定模型目录，包含 GLM-5.3、GLM-5.2、GLM-5-Turbo、GLM-5V-Turbo、GLM-5.1 与 GLM-4.7 等模型，避免依赖套餐端点的模型列表接口。
• 参考 OpenClaw 当前实现，按每次请求的实际模型处理推理强度；请求级设置优先于 Provider 配置，旧模型不会收到其协议范围外的 reasoning_effort。
• GLM-5.3 支持 low、high、max，默认 max；GLM-5.2 保留 off、high、max 语义，并保留显式 thinking.enabled 设置。
• 流式工具调用默认发送 tool_stream=true，并覆盖工具参数分片组装测试。
• 未设置 OpenClaw 客户端标识，也未修改普通智谱 Provider 的鉴权方式。
• This is NOT a breaking change. / 这不是一个破坏性变更。

Screenshots or Test Results / 运行截图或测试结果

python -m pytest tests/test_zhipu_coding_plan_source.py tests/test_openai_source.py tests/test_fastapi_v1_dashboard.py -q
191 passed, 1 warning in 69.96s

ruff check astrbot/core/provider/sources/zhipu_coding_plan_source.py astrbot/core/provider/manager.py astrbot/core/config/default.py
All checks passed!

ruff format --check astrbot/core/provider/sources/zhipu_coding_plan_source.py tests/test_zhipu_coding_plan_source.py
2 files already formatted

另使用现有 Coding Plan 密钥执行了实际 Provider 最小请求：国内 Coding API 返回 HTTP 200，AstrBot 成功解析响应并取得 token usage；请求未配置 OpenClaw 标识。密钥和响应正文均未写入测试日志。

---

Checklist / 检查清单

• 😊 If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc.
  / 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。
• 👀 My changes have been well-tested, and "Verification Steps" and "Screenshots" have been provided above.
  / 我的更改经过了良好的测试，并已在上方提供了“验证步骤”和“运行截图”。
• 🤓 I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in requirements.txt and pyproject.toml.
  / 我确保没有引入新依赖库，或者引入新依赖库的同时将其添加到 requirements.txt 和 pyproject.toml 文件相应位置。
• 😮 My changes do not introduce malicious code.
  / 我的更改没有引入恶意代码。

Summary by Sourcery

Add a separate Zhipu GLM Coding Plan provider with model-aware request handling and streaming tool support without changing the existing Zhipu integration.

New Features:

• Add a dedicated Zhipu GLM Coding Plan provider with domestic and configurable global Coding API endpoints.
• Expose a static Coding Plan model catalog and configure GLM-5.3 as the default model.
• Support model-aware reasoning controls and streaming tool-call handling for Coding Plan requests.

Enhancements:

• Preserve the existing general Zhipu provider and authentication behavior while applying request-level reasoning settings appropriately across supported model versions.

Tests:

• Add comprehensive coverage for provider registration, endpoint and model defaults, reasoning policies, request overrides, streaming behavior, and tool-call argument assembly.

[sourcery-ai]

Hey - I've found 1 issue

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="astrbot/core/provider/sources/zhipu_coding_plan_source.py" line_range="81-82" />
<code_context>
+        provider_settings: dict,
+    ) -> None:
+        merged_provider_config = dict(provider_config)
+        merged_provider_config.setdefault("api_base", ZHIPU_CODING_PLAN_API_BASE)
+        merged_provider_config.setdefault("model", ZHIPU_CODING_PLAN_DEFAULT_MODEL)
+
+        configured_extra_body = merged_provider_config.get("custom_extra_body")
</code_context>
<issue_to_address>
**issue (bug_risk):** The Coding API defaults are applied with `setdefault`, so an explicitly present empty `api_base` or `model` is preserved. An empty `api_base` makes `AsyncOpenAI` target its ordinary OpenAI default endpoint rather than the Coding API, and an empty model produces an invalid Coding API request.

**Triggers:** When an existing or manually edited provider configuration contains `api_base: ""` or `model: ""`.

**Suggested fix:** Apply the Coding defaults when the values are missing or blank, for example with `if not merged_provider_config.get("api_base")` and the equivalent model check.

```suggestion
        if not merged_provider_config.get("api_base"):
            merged_provider_config["api_base"] = ZHIPU_CODING_PLAN_API_BASE
        if not merged_provider_config.get("model"):
            merged_provider_config["model"] = ZHIPU_CODING_PLAN_DEFAULT_MODEL
```
</issue_to_address>

Sourcery assessment

Approval pending. 1 finding to address first.

Blocking findings: astrbot/core/provider/sources/zhipu_coding_plan_source.py:82

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[sourcery-ai]

issue (bug_risk): The Coding API defaults are applied with setdefault, so an explicitly present empty api_base or model is preserved. An empty api_base makes AsyncOpenAI target its ordinary OpenAI default endpoint rather than the Coding API, and an empty model produces an invalid Coding API request.

Triggers: When an existing or manually edited provider configuration contains api_base: "" or model: "".

Suggested fix: Apply the Coding defaults when the values are missing or blank, for example with if not merged_provider_config.get("api_base") and the equivalent model check.

Suggested change

	merged_provider_config.setdefault("api_base", ZHIPU_CODING_PLAN_API_BASE)
	merged_provider_config.setdefault("model", ZHIPU_CODING_PLAN_DEFAULT_MODEL)
	if not merged_provider_config.get("api_base"):
	merged_provider_config["api_base"] = ZHIPU_CODING_PLAN_API_BASE
	if not merged_provider_config.get("model"):
	merged_provider_config["model"] = ZHIPU_CODING_PLAN_DEFAULT_MODEL