fix: WriteFile error with object content (#535)

[MrMYHuang]

TLDR

A workaround for WriteFile issue in #535.

Dive Deeper

Some original params.content type is object, but their schema type is string. My workaround checks this case and convert the object to string. I agree that this workaround is a bit ugly😅, but it just works.

Reviewer Test Plan

Testing Matrix

	🍏	🪟	🐧
npm run	❓	❓	❓
npx	❓	❓	❓
Docker	❓	❓	❓
Podman	❓	-	-
Seatbelt	❓	-	-

Linked issues / bugs

A partial fix for WriteFile issue in #535. "Edit" issue in #535 is not addressed in this PR.

[pomelo-nwu]

@MrMYHuang Thank you for your contribution! We'll need some time to review the changes and assess their potential impact, so we appreciate your patience.

If you need to use it immediately, you can test it locally via npm link. Alternatively, to share a build with others, you can run npm run build && npm run bundle && npm run prepare:package && cd dist && npm pack from the root directory. This will generate a .tgz file that you can distribute. Others can then install it globally with npm i -g qwen-code-qwen-code-xxx.tgz to give it a try.

Thanks again for your work on this. We aim to process your PR as soon as we can.

[MrMYHuang]

@MrMYHuang Thank you for your contribution! We'll need some time to review the changes and assess their potential impact, so we appreciate your patience.

If you need to use it immediately, you can test it locally via npm link. Alternatively, to share a build with others, you can run npm run build && npm run bundle && npm run prepare:package && cd dist && npm pack from the root directory. This will generate a .tgz file that you can distribute. Others can then install it globally with npm i -g qwen-code-qwen-code-xxx.tgz to give it a try.

Thanks again for your work on this. We aim to process your PR as soon as we can.

Hello,

My latest study shows this issue could be a bug of LM Studio:
#535 (comment)
llama-server has no such issue.

I also open an issue to LM Studio:
lmstudio-ai/lmstudio-bug-tracker#1191
If LM Studio confirm it's a bug of its app, I will close this PR.

In the mean time, llama-server is an alternative to LM Studio without my modified qwen-code.