fix(qwen): forward role-specific model config on launch and restore

[Ashutsoh1729]

What

Forward the configured role model to the qwen CLI on both launch and restore, and advertise the model config field.

Why

Fixes #2893.

GetLaunchCommand and GetRestoreCommand built their argv without reading cfg.Config.Model, so a worker or orchestrator configured with a specific model silently started on Qwen Code's global default. The adapter also inherited the empty agentbase.Base.GetConfigSpec, so it never advertised a model field.

How

Mirrors the Codex fix in #2869:

• appendModelFlag appends --model <trimmed model> when strings.TrimSpace(cfg.Config.Model) != "", called from both GetLaunchCommand and GetRestoreCommand.
• GetConfigSpec override advertising model as ports.ConfigFieldString.

The flag is appended right after the permission flags so it sits with the other append*Flags output, ahead of --prompt or --resume. Qwen Code gives the CLI flag the highest precedence, so this wins over a user's config file value as intended.

Model values stay in Qwen's provider/model form; the adapter passes the string through untouched apart from trimming.

Testing

cd backend && go build ./... && go test -v ./internal/adapters/agent/qwen/...

Three new tests, proven to fail on main before the fix:

--- PASS: TestGetLaunchCommandAppendsConfiguredModel (0.00s)
--- PASS: TestGetLaunchCommandOmitsBlankConfiguredModel (0.00s)
--- PASS: TestGetRestoreCommandAppendsConfiguredModel (0.00s)
--- PASS: TestGetConfigSpecReportsModelField (0.00s)

TestGetLaunchCommandOmitsBlankConfiguredModel passes both before and after, so it pins the blank/whitespace case rather than restating the fix.

Checklist

• Branched from main
• One focused change; links the related issue
• Follows AGENTS.md conventions and PR hygiene
• Tests added for user-visible behavior
• Relevant CI checks pass for the area touched

[somewherelostt]

Thanks for contributing to Agent Orchestrator.

This PR is being picked up by the current external contributor on-call pair:

• @illegalcall
• @Pulkit7070

If someone is already working on this, please continue as usual.
The on-call pair is added for visibility, tracking, and support, not to take over the work.
If you need help with review, direction, reproduction, or next steps, please tag @illegalcall and @Pulkit7070 here.

For faster context or live questions, you can also join the AO Discord.

Join the session here:
https://discord.gg/H6ZDcUXmq

Come by if you want to see what is being built, ask questions, or just hang around with the community.

[Ashutsoh1729]

Review request

This PR mirrors the model-config forwarding pattern from #2869 (Codex) and #2969 (Auggie).

What I've verified

• qwen v0.20.1 installed via brew install qwen-code — confirmed -m, --model <id> and -r, --resume <sessionId> flags exist and can coexist in the same command.
• go build ./... ✅
• go test -v ./internal/adapters/agent/qwen/... ✅ (all 25 tests pass, including 4 new ones)
• go vet ./internal/adapters/agent/qwen/... ✅

Open question: godoc for invalid-model handling

In the Auggie PR (#2969), a reviewer asked to document what happens when the model is invalid. The decision was to treat the model as an opaque value — trimmed and forwarded as-is, with the CLI validating it and surfacing the error, and no silent fallback. The author added a godoc comment on appendModelFlag to state this explicitly.

Our appendModelFlag currently has only a minimal godoc: "appendModelFlag appends --model if a non-empty model is configured." Would you like me to expand it to match the Auggie convention and document the passthrough-and-surface behavior? Happy to add it if that's the preferred convention.

Anything else?

Let me know if there are any other gaps or changes needed before this can be merged.

@Pulkit7070 @nikhilachale