[FEATURE] Add qwen-code compatibility

[Ivan951236]

Problem

It solves the problem of poor people with both a weak computer and a little to no cash in their wallet/card

Because qwen-code only requires a free Qwen account

Proposed solution

We can use the "-p" flag in the qwen executable that is present in the official "qwen-code" npm package,

Basically like this.

qwen -p "text here"

The "text here" part is replaced by the actual prompt.

Basically instead of using paid remote solutions or local solutions, we can resort to free remote solutions by Qwen.

To check for qwen-code presence, we can run this command.

which qwen

Alternatives considered

There are no viable alternatives to this, other than using local versions of Qwen coder models.

[Sahilbhatane]

We have OLLAMA PR opened #361, which uses local OLLAMA model instead of API integration, qwen models and other small LLM will be considered as Post-MVP feature, keeping this issue for future implementation 👍🏻

[Kesavaraja67]

I would like to work on this issue
/assign

[Anshgrover23]

@Kesavaraja67 Not this one, can u try: #146

[divanshu-go]

@mikejmorgan-ai Could you assign this to me? You had asked me about it on Discord:
https://discord.com/channels/1436619978227843182/1436619979222024314/1456448170929422346

[Suyashd999]

We have OLLAMA PR opened #361, which uses local OLLAMA model instead of API integration, qwen models and other small LLM will be considered as Post-MVP feature, keeping this issue for future implementation 👍🏻

@divanshu-go

[divanshu-go]

@Suyashd999 @Sahilbhatane

• I was explicitly asked by Mike Morgan to take this up on Discord.
  Reference: https://discord.com/channels/1436619978227843182/1436619979222024314/1456448170929422346

• This issue is not about running Qwen models locally (Ollama).

  The plan is to use Qwen’s official OAuth device authorization flow to obtain user tokens and make requests to Qwen’s free remote model completion API (e.g. https://chat.qwen.ai/api/v1/...), so users don’t need to run local models.