MLBEDSW-9892: MLCE: SOFTMAX int16 produces output diff

 - Softmax int16 should use double precision in ElementwiseMulScale
in order to match reference and avoid rounding errors
 - Update ElementwiseMulScale to also support double, default
behavior is float

Change-Id: I207f95c273c66d99d994df2aa9cde144a01f80fe
Signed-off-by: Johan Alfven <johan.alfven@arm.com>