Byteorder mark removal in UncSource removes first left angle bracket '<' of xml document

[sonntam]

For me loading of the update XML file fails because the byteorder mark removal in the UncSource.GetUpdatesFeed() method doesn't work as expected.

I found a solution at StackOverflow that fixes the issue.
The File.ReadAllText() method should recognize the UTF coding automatically anyway (see MSDN).

I implemented the fix in my fork. I suggest cherry-picking the commit sonntam/NAppUpdate@4522d37.

[synhershko]

Can you please submit a PR with a test checking several files in several formats?

[sonntam]

I'm on it.