[FEA]: Improve the RSSSourceStage for Sherlock workflows

[mdemoret-nv]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

High

Please provide a clear description of problem this feature solves

There are a few missing features that are necessary to support Sherlock workflows using RSS feeds as a source.

Describe your ideal solution

The following features should be added to the RSSSourceStage:

1. Ability to specify more than one URL to fetch as a source
2. Ability to manually specify whether it should run indefinitely or only once
3. Ability to manually specify the batch size different from the pipeline batch size
4. Ability to cache the results using requests_cache (for testing and to prevent getting blocked for too many requests)
5. Improve the parsing capability to use different sources if one fails
6. Not sure why, but sometimes using requests to download the feed and then parsing it with feedparser works better. Add a fallback path when a parsing error occurs to use a secondary method before erroring.
7. Add ability to continue even when parsing fails.

Additional context

See prototype implementation of the above changes here: https://github.com/nv-morpheus/Morpheus/blob/1f06493e3d9fe6ec22c0b373f6284cc8369d6f52/morpheus/controllers/rss_controller.py

Code of Conduct

• I agree to follow this project's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request

[mdemoret-nv]

Fixed by #1285