Officefile detection support

[Qnerd]

Detection of Office-Files based on this article:
http://serverfault.com/questions/338087/making-libmagic-file-detect-docx-files/377792#377792

Some functions may be improved and the mimetypes are hardcoded atm (should be moved to magic.xml?).

[arimus]

It looks like the tests aren't passing on this quite yet. I'm getting this on the Excel file test:

---

Test set: net.sf.jmimemagic.MagicTest

Tests run: 26, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 0.658 sec <<< FAILURE!
testXlsx(net.sf.jmimemagic.MagicTest) Time elapsed: 0.006 sec <<< FAILURE!
junit.framework.AssertionFailedError: error in testXlsx(). message: expected:<...spreadsheetml.shee...> but was:<...wordprocessingml.documen...>
at junit.framework.Assert.fail(Assert.java:47)
at net.sf.jmimemagic.MagicTest.testXlsx(MagicTest.java:564)

[Qnerd]

It seems that this modification clashes with an other "Office-Document"-Recognition change.

The problem with the actual solution is, that an file extension is needed.

I need to get the right match from an byte-array, so there is no extension available. In that case, the library only matches the content as a zip file.