[Android] System.Tests.DateTimeOffsetTests & DateTimeOffset test failures

[steveisok]

I believe these test failures are due to differences in Android's handling of the th-TH culture, but we should validate fully.

1.) DateTimeOffsetTests.ToString_MatchesExpected & TryFormat_MatchesExpected

Several failures surrounding:

runtime/src/libraries/System.Runtime/tests/System/DateTimeOffsetTests.cs

Line 1245 in 57bfe47

yield return new object[] { DateTimeOffset.MaxValue, "yy", thTH, "42" };

runtime/src/libraries/System.Runtime/tests/System/DateTimeOffsetTests.cs

Lines 1251 to 1252 in 57bfe47

	if (PlatformDetection.IsNotInvariantGlobalization)
	yield return new object[] { DateTimeOffset.MaxValue, new string('y', i), thTH, 10542.ToString("D" + i) };

Example failures:

Assert.Equal() Failure
              ↓ (pos 0)
    Expected: 42
    Actual:   99

Expected: 0010542
    Actual:   0009999

2.) DateTimeOffsetTests.ToString_Invoke_ReturnsExpected & TryFormat_MatchesExpected

Same problem as #1 with test failures surrounding:

runtime/src/libraries/System.Runtime/tests/System/DateTimeTests.cs

Line 2044 in 57bfe47

yield return new object[] { DateTime.MaxValue, "yy", thTH, "42" };

runtime/src/libraries/System.Runtime/tests/System/DateTimeTests.cs

Line 2048 in 57bfe47

yield return new object[] { DateTime.MaxValue, new string('y', i), thTH, 10542.ToString("D" + i) };

[dotnet-issue-labeler]

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

[jeffhandley]

Tagging @dotnet/area-system-runtime since the bot didn't.

[mkhamoyan]

cc @vitek-karas

[vitek-karas]

/cc @matouskozak

[Clockwork-Muse]

... these tests are relying on (loaded) host culture data, which isn't guaranteed to be stable (we've had multiple issues from when ICU decided to add non-breaking spaces, for instance).
Should we manually construct the relevant culture details?