Create node visitor for each type of document to process

[DaveJarvis]

The MainWindow::spellcheck method performs a spellcheck presuming that the document is Markdown. This isn't always the case: sometimes the document is XML-based. This currently causes no harm because the spellcheck functionality will not operate on XML documents.

Resolving this issue as follows:

• Create a text visitor interface that can, given a document (either complete or partial), extract the textual content from each node.
• Implement the visitor interface for Markdown, R Markdown, XML, and R XML.
• Instantiate the visitor that corresponds to the file type being edited.
• Remove the hard-coded parser from MainWindow.
• Remove the TextVisitor class from MainWindow.
• Reuse the Markdown Parser from the processor for the Markdown-specific TextVisitor.