Create a CustomJSONDecoder wrapper, and replace swift-concurrency with swift-atomics

[ninjadev64]

Description

This PR creates a CustomJSONDecoder struct, which wraps around the ZippyJSONDecoder and JSONDecoder decode method depending on platform. ZippyJSONDecoder is used on all platforms except Linux, where it is not currently supported.

Additionally, this PR replaces swift-concurrency with swift-atomics, which builds on Linux. Thanks to LeoDog896 for originally making this change.

This should warrant an update to #113.

Type of change

• [x] New feature (non-breaking change which adds functionality)

Checklist:

• [x] My code follows the style guidelines of this project
• [x] I have performed a self-review of my own code
• [x] I have commented my code where required, particularly in hard-to-understand areas
• [x] I have made corresponding changes to the documentation comments
• [x] My changes generate no new warnings

[b-ncMN]

Please solve these :

• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:430
• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:592
• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:709

[b-ncMN]

It would also be nice to either make all of the commits build successfully you can do that by either : rebasing and x commit into y commit or do it by rewriting the history with interactive rebasing

[ninjadev64]

Please solve these :

• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:430
• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:592
• https://github.com/stackotter/delta-client/actions/runs/3080699453/jobs/4978356481#step:4:709

These linter violations are not caused by my code.

It would also be nice to either make all of the commits build successfully you can do that by either : rebasing and x commit into y commit or do it by rewriting the history with interactive rebasing

A little bit unnecessary, no?

[b-ncMN]

It would also be nice to either make all of the commits build successfully you can do that by either : rebasing and x commit into y commit or do it by rewriting the history with interactive rebasing

A little bit unnecessary, no?

Do you really want to merge an unclean history into main?????

[ninjadev64]

Do you really want to merge an unclean history into main?????

It's not like main has a 100% clean track record, but stackotter should squash this PR before merging anyway.

[b-ncMN]

It would also be nice to either make all of the commits build successfully you can do that by either : rebasing and x commit into y commit or do it by rewriting the history with interactive rebasing

A little bit unnecessary, no?

Do you really want to merge an unclean history into main?????

It's not like main has a 100% clean track record, but stackotter should squash this PR before merging anyway.

this makes it all even worse, god damn I give up

[stackotter]

@InfRandomness it is nice to have a clean commit history on main, I agree. However, I don't want to force contributors to rebase cause not everyone knows how to, so I usually just squash merge if the commits don't obviously need to be separate. Mostly PRs are pretty small so it's not sacrificing too much commit granularity.