Reland "[scanner] Simplify TemplateSpan raw literal character handling"

This is a reland of c1226cea1ec11d5b766226c912c475647a731274

Original change's description:
> [scanner] Simplify TemplateSpan raw literal character handling
> 
> Instead of adding and removing literal chars, only add raw literal characters when we have to and never remove them.
> 
> Change-Id: Ib604c8c9fb69a96708eec3a03de102e0668c01d7
> Reviewed-on: https://chromium-review.googlesource.com/1167505
> Reviewed-by: Marja Hölttä <marja@chromium.org>
> Reviewed-by: Caitlin Potter <caitp@igalia.com>
> Commit-Queue: Toon Verwaest <verwaest@chromium.org>

Change-Id: Ia15501d75c3beaf336e90a80e0abb738f696ef9e
Reviewed-on: https://chromium-review.googlesource.com/1170604Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55040}

src/parsing/scanner.cc

@@ -1059,30 +1059,27 @@ Token::Value Scanner::ScanTemplateSpan() {
   const bool in_template_literal = true;
   while (true) {
     uc32 c = c0_;
-    Advance<capture_raw>();
+    Advance();
     if (c == '`') {
       result = Token::TEMPLATE_TAIL;
-      ReduceRawLiteralLength(1);
       break;
     } else if (c == '$' && c0_ == '{') {
-      Advance<capture_raw>();  // Consume '{'
-      ReduceRawLiteralLength(2);
+      Advance();  // Consume '{'
       break;
     } else if (c == '\\') {
       DCHECK(!unibrow::IsLineTerminator(kEndOfInput));
+      if (capture_raw) AddRawLiteralChar('\\');
       if (unibrow::IsLineTerminator(c0_)) {
         // The TV of LineContinuation :: \ LineTerminatorSequence is the empty
         // code unit sequence.
         uc32 lastChar = c0_;
-        Advance<capture_raw>();
+        Advance();
         if (lastChar == '\r') {
-          ReduceRawLiteralLength(1);  // Remove \r
-          if (c0_ == '\n') {
-            Advance<capture_raw>();  // Adds \n
-          } else {
-            AddRawLiteralChar('\n');
-          }
+          // Also skip \n.
+          if (c0_ == '\n') Advance();
+          lastChar = '\n';
         }
+        if (capture_raw) AddRawLiteralChar(lastChar);
       } else {
         bool success = ScanEscape<capture_raw, in_template_literal>();
         USE(success);
@@ -1101,14 +1098,10 @@ Token::Value Scanner::ScanTemplateSpan() {
       // The TRV of LineTerminatorSequence :: <CR><LF> is the sequence
       // consisting of the CV 0x000A.
       if (c == '\r') {
-        ReduceRawLiteralLength(1);  // Remove \r
-        if (c0_ == '\n') {
-          Advance<capture_raw>();  // Adds \n
-        } else {
-          AddRawLiteralChar('\n');
-        }
+        if (c0_ == '\n') Advance();  // Skip \n
         c = '\n';
       }
+      if (capture_raw) AddRawLiteralChar(c);
       AddLiteralChar(c);
     }
   }

src/parsing/scanner.h

@@ -311,10 +311,6 @@ class Scanner {
 
     int length() const { return is_one_byte_ ? position_ : (position_ >> 1); }
 
-    void ReduceLength(int delta) {
-      position_ -= delta * (is_one_byte_ ? kOneByteSize : kUC16Size);
-    }
-
     void Reset() {
       position_ = 0;
       is_one_byte_ = true;
@@ -461,11 +457,6 @@ class Scanner {
     next_.raw_literal_chars->AddChar(c);
   }
 
-  V8_INLINE void ReduceRawLiteralLength(int delta) {
-    DCHECK_NOT_NULL(next_.raw_literal_chars);
-    next_.raw_literal_chars->ReduceLength(delta);
-  }
-
   // Stops scanning of a literal and drop the collected characters,
   // e.g., due to an encountered error.
   inline void DropLiteral() {