supersize: Move [clone] from name -> flags & rewrite Cluster()

tools/binary_size/libsupersize/describe.py

 # Copyright 2017 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 """Methods for converting model objects to human-readable formats."""
 
 import datetime
 import itertools
 import time
 
 import models
@@ skipping 258 matching lines @@
 def DescribeSizeInfoCoverage(size_info):
   """Yields lines describing how accurate |size_info| is."""
   for section in models.SECTION_TO_SECTION_NAME:
     if section == 'd':
       expected_size = sum(v for k, v in size_info.section_sizes.iteritems()
                           if k.startswith('.data'))
     else:
       expected_size = size_info.section_sizes[
           models.SECTION_TO_SECTION_NAME[section]]
 
-    # Use raw_symbols in case symbols contains groups.
-    in_section = models.SymbolGroup(size_info.raw_symbols).WhereInSection(
-        section)
+    in_section = size_info.raw_symbols.WhereInSection(section)
     actual_size = in_section.size
     size_percent = _Divide(actual_size, expected_size)
     yield ('Section {}: has {:.1%} of {} bytes accounted for from '
            '{} symbols. {} bytes are unaccounted for.').format(
                section, size_percent, actual_size, len(in_section),
                expected_size - actual_size)
     star_syms = in_section.WhereNameMatches(r'^\*')
     padding = in_section.padding - star_syms.padding
     anonymous_syms = star_syms.Inverted().WhereHasAnyAttribution().Inverted()
     yield '* Padding accounts for {} bytes ({:.1%})'.format(
@@ skipping 49 matching lines @@
 
 def GenerateLines(obj, verbose=False, recursive=False):
   """Returns an iterable of lines (without \n) that describes |obj|."""
   return Describer(verbose=verbose, recursive=recursive).GenerateLines(obj)
 
 
 def WriteLines(lines, func):
   for l in lines:
     func(l)
     func('\n')