Deprecate String.concat with error severity.

Fix concat j2k conversion.
Add expected deprecation errors for Java String constructors that must be converted to special Kotlin String methods.

j2k/src/org/jetbrains/kotlin/j2k/SpecialMethod.kt

@@ -316,6 +316,10 @@ enum class SpecialMethod(val qualifiedClassName: String?, val methodName: String
         }
     },
 
+    STRING_CONCAT(JAVA_LANG_STRING, "concat", 1) {
+        override fun convertCall(qualifier: PsiExpression?, arguments: Array<PsiExpression>, typeArgumentsConverted: List<Type>, codeConverter: CodeConverter)
+                = BinaryExpression(codeConverter.convertExpression(qualifier), codeConverter.convertExpression(arguments.single()), Operator(JavaTokenType.PLUS).assignNoPrototype())
+    },
 
     STRING_COMPARE_TO_IGNORE_CASE(JAVA_LANG_STRING, "compareToIgnoreCase", 1) {
         override fun convertCall(qualifier: PsiExpression?, arguments: Array<PsiExpression>, typeArgumentsConverted: List<Type>, codeConverter: CodeConverter)

j2k/testData/fileOrElement/methodCallExpression/stringMethods.java

@@ -32,7 +32,6 @@ class A {
         s.codePointCount(0, s.length());
         s.offsetByCodePoints(0, 4);
         s.compareTo("test 2");
-        s.concat(" another");
         s.contains("seq");
         s.contentEquals(new StringBuilder(s));
         s.contentEquals(new StringBuffer(s));
@@ -79,6 +78,7 @@ class A {
         int limit = 5;
         useSplit(s.split("\\s+", limit));
         s.trim();
+        s.concat(" another");
 
         s.getBytes();
         s.getBytes(Charset.forName("utf-8"));

j2k/testData/fileOrElement/methodCallExpression/stringMethods.kt

@@ -1,3 +1,5 @@
+// ERROR: Using 'String(ByteArray, String): String' is an error. Use String(bytes, Charset) instead.
+// ERROR: Using 'String(ByteArray, Int, Int, String): String' is an error. Use String(bytes, offset, length, Charset) instead.
 import java.nio.charset.Charset
 import java.util.*
 
@@ -33,7 +35,6 @@ internal class A {
         s.codePointCount(0, s.length)
         s.offsetByCodePoints(0, 4)
         s.compareTo("test 2")
-        s.concat(" another")
         s.contains("seq")
         s.contentEquals(StringBuilder(s))
         s.contentEquals(StringBuffer(s))
@@ -81,6 +82,7 @@ internal class A {
         val limit = 5
         useSplit(s.split("\\s+".toRegex(), limit.coerceAtLeast(0)).toTypedArray())
         s.trim { it <= ' ' }
+        s + " another"
 
         s.toByteArray()
         s.toByteArray(Charset.forName("utf-8"))