Cherry-pick from master: AAPT2: Fix JavaDoc first sentence extraction.

The old algorithm for detecting the first sentence of a JavaDoc comment
looked for the first occurence of '.'. This does not work when code or a
{@link android.R.styleable} link is encountered in the first sentence.

Switch to checking for whitespace characters after the '.' character.

Bug: 62900335
Test: make aapt2_tests , make ds-docs
Original change-Id: I8238f6a6304c9c2f92e2e576ca8962a59c2b20ea

Change-Id: Ie272d0d793d157f2a30f3ead20b2b53976661d33
diff --git a/tools/aapt2/text/Unicode.h b/tools/aapt2/text/Unicode.h
index 2707187..546714e 100644
--- a/tools/aapt2/text/Unicode.h
+++ b/tools/aapt2/text/Unicode.h
@@ -40,6 +40,10 @@
 // characters in the ID_Continue set.
 bool IsXidContinue(char32_t codepoint);
 
+// Returns true if the Unicode codepoint has the White_Space property.
+// http://unicode.org/reports/tr44/#White_Space
+bool IsWhitespace(char32_t codepoint);
+
 // Returns true if the UTF8 string can be used as a Java identifier.
 // NOTE: This does not check against the set of reserved Java keywords.
 bool IsJavaIdentifier(const android::StringPiece& str);