[Bug] Mistral provider (zai-glm-5-2) fails with AI_TypeValidationError on streamed tool calls (@ai-sdk/mistral dependency update needed)

Description

Environment

• OpenCode Version: 1.18.29
• OS: CachyOS (Arch Linux)
• Model / Provider: Mistral (mistral/zai-glm-5-2)

Problem Description

When using Mistral models (specifically zai-glm-5-2) with tool calling enabled, streaming responses crash with an AI_TypeValidationError.

The error trace indicates that OpenCode's internal @ai-sdk/mistral adapter expects an id field on every streamed tool-call chunk, whereas incremental continuation chunks only stream partial arguments without repeating the id.

Error Log

AI_TypeValidationError: Type validation failed: 
Value: {"id":"34ab1513a87f422fbc31733708620b96","object":"chat.completion.chunk","created":1789112573,"model":"zai-glm-5-2","choices":[{"index":0,"delta":{"tool_calls":[{"type":"function","function":{"name":"","arguments":"\", \"url\": \"https"},"index":0}],"index":0,"content":""},"finish_reason":null,"logprobs":null}]}.

Error message: [
  {
    "expected": "string",
    "code": "invalid_type",
    "path": ["choices", 0, "delta", "tool_calls", 0, "id"],
    "message": "Invalid input: expected string, received undefined"
  }
]

Plugins

No response

OpenCode version

1.18.29

Steps to reproduce

1. Configure OpenCode to use native Mistral API with a valid Mistral API key.
2. Select the mistral/zai-glm-5-2 model.
3. Send a prompt that requires OpenCode to execute a tool call (e.g., "Read the contents of package.json in the current directory" or a WebFetch prompt).
4. Observe the stream failing immediately when the tool call starts, throwing TypeValidationError.

Screenshot and/or share link

https://opncd.ai/share/JQEsOcbk

Operating System

CachyOS (Arch Linux)

Terminal

xterm-kitty