@fix(data): Fix reachability debouncer causing delay on local inserts

[tylerjroach]

• PR title and description conform to Pull Request guidelines.

Issue #, if available:

Description of changes:

How did you test these changes?
(Please add a line here how the changes were tested)

Documentation update required?

• No
• Yes (Please include a PR link for the documentation update)

General Checklist

• Added Unit Tests
• Added Integration Tests
• Security oriented best practices and standards are followed (e.g. using input sanitization, principle of least privilege, etc)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[joekiller]

I still really dislike the reachability test and think it does more harm than good especially for systems that have many models, 15+, as it'll cause the entire data store to cease admitting transactions during network transitions which is kinda ridiculous for an "offline store". I didn't like it when I saw go in. To me the further work to it exemplifies why it's not a good solution. It's too much of a hammer and smashes data store just to make that sync engine run when a very reasonable restart from within would do.

[tylerjroach]

Hi @joekiller, this was put in place as we had a report that even in scenarios of no network, inserts for 100 items were taking around 25 seconds. This change improved local insertion time back to 200ms.

In it's new state, the debouncer acts like more of a cache. subscribing to the observable will immediately provide the current value, and will not transition to a new value unless the network has been in a stable state for 250ms.

I'd love to understand a bit more of your concerns here, especially "it'll cause the entire data store to cease admitting transactions during network transitions". Can you clarify this? As mentioned, this change no longer delays local insertions, which I do not believe was ever intended to begin with.