Cache clauses -> row group mapping

Create a shadow table. For `stats`, it'd be `_stats_rowgroups`.

It contains three columns:

- the clause (eg `city = 'Dawson Creek'`)
- the initial estimate, as a bitmap of rowgroups based on stats
- the actual observed rowgroups, as a bitmap

This papers over poorly sorted parquet files, at the cost of some disk
space. It makes interactive queries much more natural -- drilldown style
queries are much faster, as they can leverage work done by previous
queries.

eg 'SELECT * FROM stats WHERE city = 'Dawson Creek' and question_id >= 1935 and question_id <= 1940`
takes ~584ms on first run, but 9ms on subsequent runs.

We only create entries when the estimates don't match the actual
results.

Fixes #6

parquet/parquet_cursor.h

@@ -56,9 +56,12 @@ public:
 
   void ensureColumn(int col);
   bool isNull(int col);
+  unsigned int getNumRowGroups() const;
+  unsigned int getNumConstraints() const;
+  const Constraint& getConstraint(unsigned int i) const;
   parquet::Type::type getPhysicalType(int col);
   parquet::LogicalType::type getLogicalType(int col);
-  ParquetTable* getTable();
+  ParquetTable* getTable() const;
 
   int getInt32(int col);
   long getInt64(int col);