if (data[i] == strArr[found]) {

Shouldn't that be:

  if (data[i] == stringToFind.codePointAt(found)) {

... if you claim to use the same algorithm. That's what golang needs to do to index a UTF-8 string, right? (Comparing to golang single core version.)

That way both golang and Java will be performing proper unicode code point indexing.

Of course the optimization you made makes sense, but I think it'd be fair for the golang version to do same.