fix(ohos): use tauri::async_runtime::spawn in scheduler to fix stuck …

[SWangHash]

…second-round turn

The desktop entry pub fn run() creates a local tokio runtime that gets dropped after _run() returns on OHOS, which cancels all okio::spawn tasks including
un_outcome_handler. Once the outcome receiver is dropped, the channel becomes closed so the coordinator cannot notify the scheduler of turn completion. The scheduler then keeps a stale �ctive_turns entry, forcing subsequent submits into the queue forever.

Switching scheduler spawns to auri::async_runtime::spawn keeps them on Tauri's long-lived runtime.

Summary

Fixes #

Type and Areas

Type:

Areas:

Motivation / Impact

Verification

Reviewer Notes

Checklist

• This PR is focused and does not include secrets, temporary prompts, generated scratch files, or unrelated artifacts.
• Relevant verification is recorded above, or skipped checks are explained.
• User-facing strings, docs, and locales are updated where applicable.