Warning cleanup, part 2

[MrJul]

Another batch of warning fixes. This time focused on mobile, samples, and tests projects.

Notable changes:

• In Android, some Build.VERSION.SdkInt >= X checks have been replaced with OperatingSystem.IsAndroidVersionAtLeast.
• In Tizen, accessing some un-initialized platform properties now throws InvalidOperationException instead of NullReferenceException.

[maxkatz6]

I believe these PRs are good candidates for .git-blame-ignore-revs file.

[MrJul]

I believe these PRs are good candidates for .git-blame-ignore-revs file.

I don't know. There are many edits, but very localized, so you won't stumble upon a single block of 1000 lines of code all changed in the same PR - which is where the pain of losing years of history is.

In this PR, most edits are adding a ? or !, but there are bigger changes like complete #nullable enable in some files, or a couple of real test fixes (which should probably have been a different PR, but in the sea of fixing warnings, splitting changes wasn't that easy).

I've been very careful to not break anything, but since there some minor logic changes, I think you'd want them to appear immediately when searching for a potential bug in the future.

My opinion is that we should use git-blame-ignore-revs for big PRs that affect a lot of surrounding code at once without changing meaning, i.e. formatting, or moving code around in a way that git doesn't detect.

[robloo]

My opinion is that we should use git-blame-ignore-revs for big PRs that affect a lot of surrounding code at once without changing meaning, i.e. formatting, or moving code around in a way that git doesn't detect.

This is what I've seen in the vast majority of examples. The intended use is for massive refactoring and code formatting for the most part.