[FIX] expression: performance of search on one2many fields

We optimize the search on domains like `[('line_ids', 'in', ids)]`.
The condition is rewritten `('id', 'in', ids1)` where `ids1` is the
result of

    SELECT <many2one_field> FROM <comodel_table> WHERE id IN <ids>

The issue is that the latter potentially returns many duplicate values.
The fix consists in having as few duplicates as possible in `ids1`.

Note that domains like `[('line_ids.foo', '=', 42)]` implicitly benefit
from the optimization, as they are rewritten as the one above with

    ids = comodel.search([('foo', '=', 42)]).ids

closes odoo/odoo#43574

X-original-commit: a13c05fa5df9c98504e8e1cdb09d8d4a8d13ea91
Signed-off-by: Raphael Collet (rco) <rco@openerp.com>

odoo/osv/expression.py

@@ -441,14 +441,14 @@ def select_from_where(cr, select_field, from_table, where_field, where_ids, wher
     res = []
     if where_ids:
         if where_operator in ['<', '>', '>=', '<=']:
-            cr.execute('SELECT "%s" FROM "%s" WHERE "%s" %s %%s' % \
+            cr.execute('SELECT DISTINCT "%s" FROM "%s" WHERE "%s" %s %%s' % \
                 (select_field, from_table, where_field, where_operator),
                 (where_ids[0],))  # TODO shouldn't this be min/max(where_ids) ?
             res = [r[0] for r in cr.fetchall()]
         else:  # TODO where_operator is supposed to be 'in'? It is called with child_of...
             for i in range(0, len(where_ids), cr.IN_MAX):
                 subids = where_ids[i:i + cr.IN_MAX]
-                cr.execute('SELECT "%s" FROM "%s" WHERE "%s" IN %%s' % \
+                cr.execute('SELECT DISTINCT "%s" FROM "%s" WHERE "%s" IN %%s' % \
                     (select_field, from_table, where_field), (tuple(subids),))
                 res.extend([r[0] for r in cr.fetchall()])
     return res