doc-rst: Fix compilation of the pdf docbook

The rst2pdf tool is a very broken toolchain, with is not capable
of parsing complex documents. As such, it doesn't build the
media book, failing with:

	[ERROR] pdfbuilder.py:130 too many values to unpack

(using rst2pdf version 0.93.dev-r0 and Sphinx version 1.4.5)

So, make it build only the books we know that are safe to build.
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>

--

Btw, with the standard Sphinx version shipped on Fedora 24 (Sphinx
1.3.1), rst2pdf doesn't build even the simple kernel-documentation,
failing with this error:
    writing Kernel... [ERROR] pdfbuilder.py:130 list index out of range

This is a known bug:
    https://github.com/sphinx-doc/sphinx/issues/1844

So, maybe we should just disable pdf generation from RST for good,
as I suspect that maintaining it with a broken toolchain will be a
big headache.

Documentation/conf.py

@@ -411,7 +411,7 @@ epub_exclude_files = ['search.html']
 # multiple PDF files here actually tries to get the cross-referencing right
 # *between* PDF files.
 pdf_documents = [
-    ('index', u'Kernel', u'Kernel', u'J. Random Bozo'),
+    ('kernel-documentation', u'Kernel', u'Kernel', u'J. Random Bozo'),
 ]
 
 # kernel-doc extension configuration for running Sphinx directly (e.g. by Read