Вы можете использовать itertools.groupby
:
import itertools, sqlite3
d1 = list(sqlite3.connect('db.db').cursor().execute("SELECT id, item, price FROM table1"))
d2 = list(sqlite3.connect('db.db').cursor().execute("SELECT id, item, price FROM table2"))
full_data = sorted(d1+d2, key=lambda x:x[0])
grouped = [[a, list(b)] for a, b in itertools.groupby(full_data, key=lambda x:x[0])]
final_grouped = [[a, *[i for c in b for i in c[1:]]] for a, b in grouped]
conn = sqlite3.connect('db.db')
conn.execute('CREATE TABLE file3 (id real, item text, price real, shop real, color text)')
conn.executemany('INSERT INTO file3 VALUES (?, ?, ?, ?, ?)', final_grouped)
conn.commit()
conn.close()