few small MySQL bugs/issues that impact the engines, as discussed in the SE summit

* remove handler::index_read_last()
* create handler::keyread_read_time() (was get_index_only_read_time() in opt_range.cc)
* ha_show_status() allows engine's show_status() to fail
* remove HTON_FLUSH_AFTER_RENAME
* fix key_cmp_if_same() to work for floats and doubles
* set table->status in the server, don't force engines to do it
* increment status vars in the server, don't force engines to do it

mysql-test/r/status_user.result:
  correct test results - innodb was wrongly counting internal
  index searches as handler_read_* calls.
sql/ha_partition.cc:
  compensate for handler incrementing status counters -
  we want to count only calls to underlying engines
sql/handler.h:
  inline methods moved to sql_class.h
sql/key.cc:
  simplify the check
sql/opt_range.cc:
  move get_index_only_read_time to the handler class
sql/sp.cc:
  don't use a key that's stored in the record buffer -
  the engine can overwrite the buffer with anything, destroying the key
sql/sql_class.h:
  inline handler methods that need to see THD and TABLE definitions
sql/sql_select.cc:
  no ha_index_read_last_map anymore
sql/sql_table.cc:
  remove HTON_FLUSH_AFTER_RENAME
sql/table.cc:
  set HA_CAN_MEMCMP as appropriate
sql/tztime.cc:
  don't use a key that's stored in the record buffer -
  the engine can overwrite the buffer with anything, destroying the key
storage/myisam/ha_myisam.cc:
  engines don't need to update table->status or use ha_statistic_increment anymore
storage/myisam/ha_myisam.h:
  index_read_last_map is no more

sql/opt_range.cc

@@ -707,10 +707,7 @@ TRP_ROR_INTERSECT *get_best_covering_ror_intersect(PARAM *param,
 static
 TABLE_READ_PLAN *get_best_disjunct_quick(PARAM *param, SEL_IMERGE *imerge,
                                          double read_time);
-static
-TRP_GROUP_MIN_MAX *get_best_group_min_max(PARAM *param, SEL_TREE *tree);
-static double get_index_only_read_time(const PARAM* param, ha_rows records,
-                                       int keynr);
+static TRP_GROUP_MIN_MAX *get_best_group_min_max(PARAM *param, SEL_TREE *tree);
 
 #ifndef DBUG_OFF
 static void print_sel_tree(PARAM *param, SEL_TREE *tree, key_map *tree_map,
@@ -2314,9 +2311,9 @@ int SQL_SELECT::test_quick_select(THD *thd, key_map keys_to_use,
     if (!head->covering_keys.is_clear_all())
     {
       int key_for_use= find_shortest_key(head, &head->covering_keys);
-      double key_read_time= (get_index_only_read_time(&param, records,
-                                                     key_for_use) +
-                             (double) records / TIME_FOR_COMPARE);
+      double key_read_time= param.table->file->keyread_read_time(key_for_use,
+                                                                 1, records) +
+                             (double) records / TIME_FOR_COMPARE;
       DBUG_PRINT("info",  ("'all'+'using index' scan will be using key %d, "
                            "read time %g", key_for_use, key_read_time));
       if (key_read_time < read_time)
@@ -3938,42 +3935,6 @@ skip_to_ror_scan:
 }
 
 
-/*
-  Calculate cost of 'index only' scan for given index and number of records.
-
-  SYNOPSIS
-    get_index_only_read_time()
-      param    parameters structure
-      records  #of records to read
-      keynr    key to read
-
-  NOTES
-    It is assumed that we will read trough the whole key range and that all
-    key blocks are half full (normally things are much better). It is also
-    assumed that each time we read the next key from the index, the handler
-    performs a random seek, thus the cost is proportional to the number of
-    blocks read.
-
-  TODO:
-    Move this to handler->read_time() by adding a flag 'index-only-read' to
-    this call. The reason for doing this is that the current function doesn't
-    handle the case when the row is stored in the b-tree (like in innodb
-    clustered index)
-*/
-
-static double get_index_only_read_time(const PARAM* param, ha_rows records,
-                                       int keynr)
-{
-  double read_time;
-  uint keys_per_block= (param->table->file->stats.block_size/2/
-			(param->table->key_info[keynr].key_length+
-			 param->table->file->ref_length) + 1);
-  read_time=((double) (records+keys_per_block-1)/
-             (double) keys_per_block);
-  return read_time;
-}
-
-
 typedef struct st_ror_scan_info
 {
   uint      idx;      /* # of used key in param->keys */
@@ -4050,8 +4011,8 @@ ROR_SCAN_INFO *make_ror_scan(const PARAM *param, int idx, SEL_ARG *sel_arg)
       bitmap_set_bit(&ror_scan->covered_fields, key_part->fieldnr-1);
   }
   ror_scan->index_read_cost=
-    get_index_only_read_time(param, param->table->quick_rows[ror_scan->keynr],
-                             ror_scan->keynr);
+    param->table->file->keyread_read_time(ror_scan->keynr, 1,
+                                    param->table->quick_rows[ror_scan->keynr]);
   DBUG_RETURN(ror_scan);
 }
 
@@ -4886,7 +4847,8 @@ static TRP_RANGE *get_key_scans_params(PARAM *param, SEL_TREE *tree,
           We can resolve this by only reading through this key. 
           0.01 is added to avoid races between range and 'index' scan.
         */
-        found_read_time= get_index_only_read_time(param,found_records,keynr) +
+        found_read_time= param->table->file->keyread_read_time(keynr,1,
+                                                               found_records) +
                          cpu_cost + 0.01;
       }
       else