Add logical structure extractor from content information

[oksidgy]

Need to add logical structure extractor that analyzes a content. An extractor will analyze a content (extracted text lines) and assign a priority to each text line. So we can understand each line is a structure element or plain text. As a result we build a document tree based on text lines' priority.