add priority to Bach, Job, and run_async

[QingengWei]

Greptile Summary

This PR adds priority parameter support to the Tidy3D web API's batch processing system, specifically targeting vGPU queue management. The changes enable users to specify task priority levels from 1 (lowest) to 10 (highest) when submitting simulations to the vGPU queue.

The implementation adds the priority parameter to three key components:

1. Job class (container.py): Both Job.run() and Job.start() methods now accept an optional priority parameter
2. Batch class (container.py): Both Batch.run() and Batch.start() methods now support priority, with proper threading implementation to ensure all jobs in a batch inherit the same priority
3. run_async function (asynchronous.py): The high-level async interface now accepts priority and passes it through to the underlying batch processing

The parameter propagation follows a clear chain: run_async() → Batch.run() → Batch.start() → individual Job.start() → webapi.start(). All methods maintain backward compatibility by making priority optional with a default value of None. The change integrates well with the existing codebase architecture and follows established patterns for parameter passing through the web API layers.

This enhancement gives users fine-grained control over task execution order in the vGPU queue, which is valuable for managing multiple simulations with different urgency levels.

Important Files Changed

Click to expand file changes

Filename	Score	Overview
tidy3d/web/api/container.py	4/5	Added priority parameter to Job and Batch classes with proper propagation through run/start methods
tidy3d/web/api/asynchronous.py	4/5	Added priority parameter to run_async function with documentation and batch integration

Confidence score: 4/5

• This PR is generally safe to merge with well-structured changes that maintain backward compatibility
• Score reflects solid implementation following established patterns, though some edge cases around parameter validation could be addressed
• Pay close attention to the threading implementation in Batch.start() to ensure priority is properly propagated to all jobs

Sequence Diagram

sequenceDiagram
    participant User
    participant run_async
    participant Batch
    participant Job
    participant web
    
    User->>run_async: "run_async(simulations, priority=5)"
    run_async->>Batch: "Batch(simulations, folder_name, ...)"
    run_async->>Batch: "batch.run(path_dir, priority)"
    Batch->>Batch: "upload()"
    Batch->>Batch: "start(priority)"
    loop for each job in batch
        Batch->>Job: "job.start(priority)"
        Job->>web: "web.start(task_id, solver_version, pay_type, priority)"
    end
    Batch->>Batch: "monitor()"
    Batch->>User: "return BatchData"

Loading

[QingengWei]

@e-g-melo @momchil-flex @tylerflex The pr is for https://flow360.atlassian.net/browse/SCEM-10928

[greptile-apps]

_{2 files reviewed, 6 comments}

_{Edit Code Review Bot Settings | Greptile}

[github-actions]

Diff Coverage

Diff: origin/develop...HEAD, staged and unstaged changes

• tidy3d/web/api/asynchronous.py (100%)
• tidy3d/web/api/container.py (100%)

Summary

• Total: 9 lines
• Missing: 0 lines
• Coverage: 100%