ICMLFilter: forced line break character should get extracted as placeholder

Created by: Martin Wunderlich

The forced line break character U+2028 is not treated in any special way when processing the text (in method ICMLFilter.processText()). This means that this character is simply appended to the text fragment as plain character and, consequently, might get overwritten by the translator, especially if the editor application doesn't render this characater as a line break.

This character shoudl get replaced by a placeholder tag, same as many other characters in the ICML filter.

See discussion here: https://groups.google.com/forum/#!topic/okapi-devel/pPTVMddvmyM