This caused Item_windowfunc to return garbage.

Query result had 0 where it should have had NULLs:
- Make Item_window_func::val* functions honor NULL-handling conventions:
  1. set null_value to indicate whether we've returned a NULL value
  2. val_str and val_decimal should return NULL pointer when they're
     returning SQL NULL value.

Fix assertion failure when sending results to network.
- The assert was due to window func returing SQL NULL despite having
  maybe_null=false
- Fixed by settting Item_window_func::maybe_null correctly in fix_fields

Make query pushdown work in the post-refactored code.
This fixes sequence.group_by test.

The counter keeps track of the number of items added to the sum
function. It is increased when we add a value to the sum function and
decreased when it is removed.

This implementation does not deal with the case where removal of
elements from the window frame causes the item to turn to a null value.

- Handle ORDER BY DESC in window definitions.
- Fix an issue in Frame_range_current_row_top

Add support for "RANGE n PRECEDING|FOLLOWING" frame bounds.
- n is currently limited to whatever Item and Item_sum_plus/minus
  can handle (i.e. no DATETIME intervals).
- Didn't check NULL value handling yet.

Also added win_bit.result (result values checked)

Part#2: Fix a couple more issues in rows-type frames.
This also has a code cleanup:
- introduce a separate Frame_rows_current_row_(top,bottom). This is
  is a special case which doesn't need its cursor or partition bound check
- Split Frame_n_rows into
  = Frame_n_rows_preceding (this one is now much simpler)
  = Frame_n_rows_following (simpler and works but may need some work still)

Part#1: Frame_n_rows::next_partition() should not assume that the
current table->record[0] points to the first row in the partition.

Since cursor supports move_to() operation, we dont need this.

The results for this test case are blocked by MDEV-9695.

Add testcases for RANGE UNBOUNDED PRECEDING|FOLLOWING

Support RANGE ... CURRENT ROW as frame's first and second bound.

Introduce explicitly-defined cursors.
- Rowid_seq_cursor just reads the rowid sequence
- Table_read_cursor also reads rows for the rowids
- Also, we need a kind of cursor that stops at partition boundary?

... into Frame_unbounded_preceding and Frame_unbounded_following

- Item_sum_count::remove() should check if the argument's value is NULL.
- Window Function item must have its Item_window_func::split_sum_func
  called,
- and it must call split_sum_func for aggregate's arguments (see the
  comment near Item_window_func::split_sum_func for explanation why)

Resolved window names.
Checked some constraints for window frames.
Added test cases for window name resolution.

- Add temporary code: clone_read_record() clones READ_RECORD structure,
  as long as it is used for reading filesort() result that fits into
  memory.
- Add frame bounds for ROWS-type frames. ROWS n PRECEDING|FOLLOWING,
  ROWS UNBOUNDED PRECEDING|FOLLOWING, CURRENT ROW are supported.
- Add Item_sum_count::remove() which allows "streaming" computation
  of COUNT() over a moving frame.

- Make queries that use multiple window functions not to leak memory
- Code cleanup in sql_window.cc

These do not have any meaning after MDEV-8646. Their only valid values are
- table_access_tabs= join_tab;
- top_table_access_tabs_count= top_join_tab_count;