Default to an Encoding without BOM when BOM marker was not found

[mdonatas]

OpenStream methods attempts detecting encoding by checking for BOM at the beginning of the stream. When that fails a call to AutoDetect is made. Given that AutoDetect is entered knowing that BOM was not found, an Encoding of StreamReader that it uses should be without BOM.
This doesn't impact reading the file but it does impact the receiving/consuming end of a StreamReader where a code might check what encoding is set on a given StreamReader to use the same encoding for saving edited file which would lead to incorrectly adding BOM.

StreamReader constructor defaults to Encoding.UTF8 which has BOM thus encoding used to create a file doesn't match encoding set on the StreamReader

public StreamReader(Stream stream, bool detectEncodingFromByteOrderMarks)
    : this(stream, Encoding.UTF8, detectEncodingFromByteOrderMarks, DefaultBufferSize, false)
{
}

[mdonatas]

p.s. this is a part of a fix for gitextensions/gitextensions#11934

[mstv]

LGTM, have not run