TEST: Mergeable Sample

[denrase]

📜 Description

I have added the mergeable flags to our build script and tested release builds in the Simulator on macOS.

The sample project is embedding the Sentry dynamic library.

Size

The archive is slightly smaller with mergeable enabled, but the installed app should be slightly larger because of additional metadata.

Default	Mergeable

Launch Time

Load times are similar. Launched both versions 10 times after the initial installation. Launching from Xcode takes longer when using mergeable instead of the default dylib.

Default	Mergeable

Conclusion

We could add the linker flags to our build script for Sentry-Dynamic. Users can then decide if they want to use mergeable. This is opt-in for them, as the default is the same as before.

We can delete this PR afterwards.

💡 Motivation and Context

Sample to test mergeable option.

Relates to #3719

[kahest]

The archive is slightly smaller with mergeable enabled, but the installed app should be slightly larger because of additional metadata.

Not sure I fully understand this - I see a difference of 1.6 MB vs. 116 KB, which is significant, but you're saying in a real-world installed app on a device it would be more than 116 KB?

[denrase]

@kahest The non-mergable has the frameworks folder, where the sentry framework is included (the folder size is not shown in the screenshot). The mergable has the sentry framework plus some metadata, so it should be slightly larger, at least thats how i understood the documentation

[brustolin]

We can delete this PR afterwards

Can we make the changes to enable Mergeable Samples instead?
If it is easier to just create a new PR that's ok.

[denrase]

Closing in favor of #4367