Implements Compressed Node Based Graph (De-)Serialization Skeleton

Implements parallel recursion for the partitioner
Fixes osrm-extract's -dump-partition-graph: accept no further tokens

References:
- http://www.boost.org/doc/libs/1_55_0/doc/html/boost/program_options/bool_switch.html

Pulls parameters through to make them configurable from the outside

Defaults are equivalent to:

    ./osrm-partition \
      berlin-latest.osrm \
      --max-cell-size 4096 \
      --balance 1.2 \
      --boundary 0.25 \
      --optimizing-cuts 10

Fixes parallel_do call for Intel TBB 4.2 (Trusty): no range-based overload

src/tools/extract.cpp

@@ -46,13 +46,18 @@ return_code parseArguments(int argc, char *argv[], extractor::ExtractorConfig &e
             ->implicit_value(true)
             ->default_value(false),
         "Generate a lookup table for internal edge-expanded-edge IDs to OSM node pairs")(
+        "dump-partition-graph",
+        boost::program_options::bool_switch(&extractor_config.dump_compressed_node_based_graph)
+            ->implicit_value(true)
+            ->default_value(false),
+        "Generate a partitionable graph file (.cnbg) for use with osrm-partition")(
         "small-component-size",
         boost::program_options::value<unsigned int>(&extractor_config.small_component_size)
             ->default_value(1000),
         "Number of nodes required before a strongly-connected-componennt is considered big "
         "(affects nearest neighbor snapping)")(
         "with-osm-metadata",
-        boost::program_options::value<bool>(&extractor_config.use_metadata)
+        boost::program_options::bool_switch(&extractor_config.use_metadata)
             ->implicit_value(true)
             ->default_value(false),
         "Use metada during osm parsing (This can affect the extraction performance).");

src/tools/partition.cpp

@@ -28,11 +28,31 @@ return_code parseArguments(int argc, char *argv[], partition::PartitionConfig &p
 
     // declare a group of options that will be allowed both on command line
     boost::program_options::options_description config_options("Configuration");
-    config_options.add_options()(
-        "threads,t",
-        boost::program_options::value<unsigned int>(&partition_config.requested_num_threads)
-            ->default_value(tbb::task_scheduler_init::default_num_threads()),
-        "Number of threads to use");
+    config_options.add_options()
+        //
+        ("threads,t",
+         boost::program_options::value<unsigned int>(&partition_config.requested_num_threads)
+             ->default_value(tbb::task_scheduler_init::default_num_threads()),
+         "Number of threads to use")
+        //
+        ("max-cell-size",
+         boost::program_options::value<std::size_t>(&partition_config.maximum_cell_size)
+             ->default_value(4096),
+         "Bisection termination citerion based on cell size")
+        //
+        ("balance",
+         boost::program_options::value<double>(&partition_config.balance)->default_value(1.2),
+         "Balance for left and right side in single bisection")
+        //
+        ("boundary",
+         boost::program_options::value<double>(&partition_config.boundary_factor)
+             ->default_value(0.25),
+         "Percentage of embedded nodes to contract as sources and sinks")
+        //
+        ("optimizing-cuts",
+         boost::program_options::value<std::size_t>(&partition_config.num_optimizing_cuts)
+             ->default_value(10),
+         "Number of cuts to use for optimizing a single bisection");
 
     // hidden options, will be allowed on command line, but will not be
     // shown to the user