[Sitemaps] Parser assumes file encoding is the same as the one used by default on JVM

SiteMapParser creates Readers from the byte[] content by assuming that the file 
encoding is the same as the one use by default on the JVM. 

We should use Tika's charset detector (or ICU4J directly) to detect the charset 
used prior to creating the Readers.

Will attach a test class shortly.

Original issue reported on code.google.com by digitalpebble on 19 Jan 2015 at 4:51

[jnioche]

@kkrugler this has been covered elsewhere indeed.

[kkrugler]

@jnioche - since you originally opened this issue, can you confirm what Sebastian is seeing and then close it if this has been subsumed by #67 and #137? Thanks!