Incorrect decoding near EOF for stateful decoders like UTF-16

ADDITIONAL SYSTEM INFORMATION :
Reproduced with 1.8.0_74, 11.0.14, 17.0.3 and 19-ea+34 on Ubuntu 18.04.

A DESCRIPTION OF THE PROBLEM :
StreamDecoder.implRead() resets the decoder at [1] after seeing the EOF, but before doing the final decoding round, so any stateful decoder will use its initial state when decoding the remaining input, which can lead to a wrong result.

The reproducer demonstrates the bug by making the UTF-16 decoder forget the autodetected BOM before decoding the final two bytes.

[1] https://github.com/openjdk/jdk/blob/ae52053757ca50c4b56989c9b0c6890e504e4088/src/java.base/share/classes/sun/nio/cs/StreamDecoder.java#L381

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
Compile and run the following code.

EXPECTED VERSUS ACTUAL BEHAVIOR :
EXPECTED -
Exception in thread "main" java.nio.charset.MalformedInputException: Input length = 2
        at java.base/java.nio.charset.CoderResult.throwException(CoderResult.java:274)
        at java.base/sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:339)
        at java.base/sun.nio.cs.StreamDecoder.read(StreamDecoder.java:178)
        at java.base/sun.nio.cs.StreamDecoder.read0(StreamDecoder.java:127)
        at java.base/sun.nio.cs.StreamDecoder.read(StreamDecoder.java:112)
        at java.base/java.io.InputStreamReader.read(InputStreamReader.java:164)
        at Test.main(Test.java:16)

ACTUAL -
216
-1
65279
Exception in thread "main" java.nio.charset.MalformedInputException: Input length = 2
        at java.base/java.nio.charset.CoderResult.throwException(CoderResult.java:274)
        at java.base/sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:339)
        at java.base/sun.nio.cs.StreamDecoder.read(StreamDecoder.java:178)
        at java.base/sun.nio.cs.StreamDecoder.read0(StreamDecoder.java:127)
        at java.base/sun.nio.cs.StreamDecoder.read(StreamDecoder.java:112)
        at java.base/java.io.InputStreamReader.read(InputStreamReader.java:164)
        at Test.main(Test.java:23)


---------- BEGIN SOURCE ----------
import java.io.*;
import java.nio.charset.Charset;
import java.nio.charset.CodingErrorAction;

public class Test {
    public static void main(String[] args) throws IOException {
        byte[] input = {
            (byte) 0xff, (byte) 0xfe, // BOM (in UTF-16LE)
            0, (byte) 0xd8, // High surrogate (in UTF-16LE)
        };
        InputStreamReader r;

        r = new InputStreamReader(
                new ByteArrayInputStream(input),
                Charset.forName("UTF-16").newDecoder().onMalformedInput(CodingErrorAction.REPORT));
        System.out.println(r.read()); // \u00d8 (wrong, uses UTF-16BE)
        System.out.println(r.read()); // EOF

        r = new InputStreamReader(
                new ByteArrayInputStream(input),
                Charset.forName("UTF-16LE").newDecoder().onMalformedInput(CodingErrorAction.REPORT));
        System.out.println(r.read()); // BOM
        System.out.println(r.read()); // MalformedInputException (correct)
    }
}
---------- END SOURCE ----------

FREQUENCY : always