Native Support for Saving Channel Recordings Directly to AWS S3 Bucket with Time-Based Folder Structure

[flowersofpepper]

- Is your feature request related to a problem? Please describe.

I'm currently using Restreamer to manage and record video streams. While it offers great flexibility for live streaming and HLS output, I’m facing difficulty automating the storage and backup of recordings directly to cloud storage. Currently, I need to manually move or sync the recordings to AWS S3, which introduces extra steps and potential delays in data availability, especially when managing multiple channels.

- Describe the solution you'd like

I would like Restreamer to natively support saving recorded video segments directly to an AWS S3 bucket. Specifically:

1. Each channel’s recordings should be saved in a corresponding folder named after the channel.
2. Inside that folder, subfolders should be created based on date and/or hour (e.g., /my-channel/2025-04-10/09/).
3. The video should be split into chunks based on a configurable time interval (e.g., every 10 minutes).
4. Optional: The ability to configure S3 credentials and target bucket directly via the UI or environment variables.

This feature would make it easy to integrate with cloud-based processing, backups, and analytics pipelines without relying on external scripts or mounts.

- Describe alternatives you've considered

1. Using a cron job or background script to sync the recordings/ folder to S3 periodically.
2. Mounting an S3 bucket as a local drive using tools like s3fs or rclone.
3. Writing a wrapper service that processes local recordings and uploads them afterward.

While these workarounds are functional, they add complexity, require more resources, and are prone to failures.

- Additional context

This feature would be especially helpful for users running Restreamer in cloud-native or containerized environments, where persistent local storage isn't ideal. It would also improve integration with modern cloud workflows (like AWS Lambda, Athena, etc.) by providing organized and time-indexed media files directly in S3.

If needed, I’d be happy to test and help iterate on this feature during development!