Aws Stackset Drift Detect

Trigger CloudFormation StackSet drift detection and record its outcome — the write-side sibling of @jentz/aws-stackset-audit. A single mutating detect method calls DetectStackSetDrift for the configured stackset, polls DescribeStackSetOperation to a terminal state (bounded by pollSeconds / maxPolls), and writes one operation resource capturing the operation id, action, terminal status, timestamps, and status reason.

It performs no instance audit and writes no summary/instance resources — that is the read-only @jentz/aws-stackset-audit. Compose the two in a swamp workflow: run this detect step first, then the audit step with dependsOn: succeeded, so the audit reads the refreshed per-instance drift status straight from AWS.

Mutating: it needs cloudformation:DetectStackSetDrift plus the stackset admin role, so a *-readonly profile deliberately cannot run it. The global-args shape (stackSetName, callAs, region, profile) is identical to the audit sibling so a workflow wires the same inputs into both steps. Only StackSet drift is in scope; stack-level DetectStackDrift is not.