Helix job monitor: surface work item failures and job errors into AzDO build timeline

[Copilot]

WaitForHelixJobCompletion was silently swallowing per-work-item failure details and logging job-level errors without a FailureCategory, meaning neither showed up as proper annotations in the AzDO build timeline.

Changes

WaitForHelixJobCompletion.cs

• Fix job-level error to use FailureCategory.Helix (was untagged Log.LogError)
• Track failed work items across polling iterations via HashSet<string>; on first detection of each failure, emit Log.LogError(FailureCategory.Test, ...) with the work item name, job name, and console log URI — same format and URI construction already used by CheckHelixJobStatus

// Report any newly failed work items so this information makes it into the AzDO build timeline.
foreach (string failedWorkItemName in pf.Failed ?? Enumerable.Empty<string>())
{
    string wi = Helpers.CleanWorkItemName(failedWorkItemName);
    if (reportedWorkItemFailures.Add(wi))
    {
        string consoleUri = HelixApi.Options.BaseUri.AbsoluteUri.TrimEnd('/') + $"/api/2019-06-17/jobs/{jobName}/workitems/{Uri.EscapeDataString(wi)}/console";
        Log.LogError(FailureCategory.Test, $"Work item {wi} in job {jobName} has failed. Failure log: {consoleUri}{accessTokenSuffix}");
    }
}

Failures are reported as they appear (each polling iteration), not deferred to CheckHelixJobStatus, giving engineers real-time visibility in the timeline without waiting for the full job to complete.

To double check:

• The right tests are in and the right validation has happened. Guidance: https://github.com/dotnet/arcade/blob/main/Documentation/Validation.md