Update benchmarks and fuzz targets bucket periodically in GKE

[Ekam219]

Hi DonggeLiu,

Thanks a lot for your guidance on this! I’ve implemented the changes to automate syncing OSS-Fuzz targets and benchmarks to GCS. For now, I’ve created a draft PR for your review. Please let me know if you’d like any changes or if you’d like to dive deeper into any part of the implementation! #66

Summary of Updates
🔹 Data Source Implementation
Added a new script: tools/gcs_sync.py

Fetches human-written fuzz targets and function signatures using target_collector.py and introspector.py

Excludes LLVMFuzzerTestOneInput

Stores timestamped JSON files in the oss-fuzz-gen-targets GCS bucket under project-specific directories

🔹 Scheduling & Automation
Created k8s/gcs-sync-cron.yaml to define a Kubernetes CronJob

Default schedule: Runs daily at 3 AM UTC (adjustable via Cron syntax)

Uses a dedicated service account (oss-fuzz-sync-sa) with minimal permissions (storage.objectAdmin) for security

🔹 GKE Integration & Docker Image
Built a lightweight Docker image (Dockerfile.sync)

Includes only essential dependencies: Python, GCS SDK, and the sync script

Optimized image size (~300MB) to prevent unnecessary bloat

Automated build & deployment via GitHub Actions

🔹 Cluster Access & Workflow
Everything is managed via YAML files and GitHub Actions

No direct GKE access required, ensuring a more secure and auditable system

Let me know if you have any feedback! 🚀
Cheers,
Ekamjot