fix: unique deployment names for cluster job

[ajbeattie]

• Ensure that the set of deployment names submitted in the cluster job are unique. For duplicates (e.g. during an active rollout), break ties by preferring running over terminal pods and preferring the newest among running pods
• Also add individual record logging for parity with the individual record posting path, and easier debugging

[Copilot]

Pull request overview

This PR improves startup “bulk cluster sync” reporting by ensuring only a single (unique) deployment_name is submitted per snapshot job, with deterministic tie-breaking during rollouts, and adds per-record INFO logging to match the single-record posting path.

Changes:

• Deduplicate startup sync records to ensure unique deployment_name values, resolving digest collisions by preferring running pods over terminal pods and otherwise choosing the newest pod.
• Add per-record INFO logging for accepted records after the cluster job is queued (excluding records rejected at submission).
• Add targeted unit tests covering rollout/tie-breaking behavior and uniqueness guarantees.

Show a summary per file

File	Description
internal/controller/reporting.go	Dedupes sync records by deployment_name, adds tie-break logic and per-record sync logging.
internal/controller/reporting_test.go	Adds regression and behavior tests for rollout collision handling and uniqueness.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 2/2 changed files
• Comments generated: 4