prevent unicode-confusion in password by applying PRECIS, and username/email address by applying unicode NFC normalization

an é (e with accent) can also be written as e+\u0301. the first form is NFC,
the second NFD. when logging in, we transform usernames (email addresses) to
NFC. so both forms will be accepted. if a client is using NFD, they can log
in too.

for passwords, we apply the PRECIS "opaquestring", which (despite the name)
transforms the value too: unicode spaces are replaced with ascii spaces. the
string is also normalized to NFC. PRECIS may reject confusing passwords when
you set a password.

vendor/golang.org/x/text/secure/precis/nickname.go

@@ -0,0 +1,72 @@
+// Copyright 2015 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package precis
+
+import (
+	"unicode"
+	"unicode/utf8"
+
+	"golang.org/x/text/transform"
+)
+
+type nickAdditionalMapping struct {
+	// TODO: This transformer needs to be stateless somehow…
+	notStart  bool
+	prevSpace bool
+}
+
+func (t *nickAdditionalMapping) Reset() {
+	t.prevSpace = false
+	t.notStart = false
+}
+
+func (t *nickAdditionalMapping) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
+	// RFC 8266 §2.1.  Rules
+	//
+	// 2.  Additional Mapping Rule: The additional mapping rule consists of
+	//     the following sub-rules.
+	//
+	//     a.  Map any instances of non-ASCII space to SPACE (U+0020); a
+	//         non-ASCII space is any Unicode code point having a general
+	//         category of "Zs", naturally with the exception of SPACE
+	//         (U+0020).  (The inclusion of only ASCII space prevents
+	//         confusion with various non-ASCII space code points, many of
+	//         which are difficult to reproduce across different input
+	//         methods.)
+	//
+	//     b.  Remove any instances of the ASCII space character at the
+	//         beginning or end of a nickname (e.g., "stpeter " is mapped to
+	//         "stpeter").
+	//
+	//     c.  Map interior sequences of more than one ASCII space character
+	//         to a single ASCII space character (e.g., "St  Peter" is
+	//         mapped to "St Peter").
+	for nSrc < len(src) {
+		r, size := utf8.DecodeRune(src[nSrc:])
+		if size == 0 { // Incomplete UTF-8 encoding
+			if !atEOF {
+				return nDst, nSrc, transform.ErrShortSrc
+			}
+			size = 1
+		}
+		if unicode.Is(unicode.Zs, r) {
+			t.prevSpace = true
+		} else {
+			if t.prevSpace && t.notStart {
+				dst[nDst] = ' '
+				nDst += 1
+			}
+			if size != copy(dst[nDst:], src[nSrc:nSrc+size]) {
+				nDst += size
+				return nDst, nSrc, transform.ErrShortDst
+			}
+			nDst += size
+			t.prevSpace = false
+			t.notStart = true
+		}
+		nSrc += size
+	}
+	return nDst, nSrc, nil
+}