MLBEDSW-1685: bug fix when IFM==IFM2 for elementwise ops

There was output diff when both IFMs are referring to the same tensor in
binary elementwise operations. IFM2 dimension-instructions were not written
by vela.

Change-Id: I40a0dcbc9557f7308222b7230e5586d8f2a04c6a
Signed-off-by: Louis Verhaard <louis.verhaard@arm.com>
diff --git a/ethosu/vela/register_command_stream_generator.py b/ethosu/vela/register_command_stream_generator.py
index c0402b3..0168c64 100644
--- a/ethosu/vela/register_command_stream_generator.py
+++ b/ethosu/vela/register_command_stream_generator.py
@@ -809,7 +809,7 @@
                         emit.cmd0_with_param(cmd0.NPU_SET_OFM_HEIGHT0_M1, height_0 - 1)
                         emit.cmd0_with_param(cmd0.NPU_SET_OFM_HEIGHT1_M1, height_1 - 1)
                         emit.cmd0_with_param(cmd0.NPU_SET_OFM_WIDTH0_M1, width_0 - 1)
-                    elif tens == cmd.ifm2_tensor:
+                    if tens == cmd.ifm2_tensor:
                         emit.cmd0_with_param(cmd0.NPU_SET_IFM2_HEIGHT0_M1, height_0 - 1)
                         emit.cmd0_with_param(cmd0.NPU_SET_IFM2_HEIGHT1_M1, height_1 - 1)
                         emit.cmd0_with_param(cmd0.NPU_SET_IFM2_WIDTH0_M1, width_0 - 1)