Inconsistent behavior for event type IGNORABLE_WHITESPACE in XmlReader

[sschirr]

Newline characters after a starting element and followed by another starting element will be detected as event type TEXT on most platform targets (native, android, js). Only on the jvm target it is (correctly) detected as a IGNORABLE_WHITESPACE.

I implemented a test for this case that is only passing on jvm target. See: #127

[pdvrieze]

The "default" implementation (rather than the platform independent implementation) uses the existing parsers for the platform (except native where the platform independent parser - based on android's one is used). I'll have a look at fixing this for the multiplatform parser.

[pdvrieze]

The parser is schemaless and technically ignorable whitespace can only be determined in the context of a schema (technically only when there can not be values is the whitespace ignorable). But the handling now marks the text as ignorable if there is no text between tags (even if there could be text after a child tag).