use rmw_enclave_options_xxx APIs instead.

[fujitatomoya]

closes #311

related to ros2/rmw#390 and ros2/rmw#390

[fujitatomoya]

@pablogs9 can you take a look?

[fujitatomoya]

@pablogs9 ros2/rmw#393 has been merged, is this good to go?

[pablogs9]

CI is ok, so we are good to go.

@fujitatomoya thanks for the contribution!

[fujitatomoya]

@pablogs9 you already reverted this PR from rolling?

[pablogs9]

Hello @fujitatomoya yes, I have reverted it temporarily because it breaks micro-ROS functionality on our hardware in the loop CI/CD, we are still investigating. As soon as I find what happens, I will restore it and open a PR to fix the problem.

[fujitatomoya]

@pablogs9 okay thanks for checking that. please be advised that this must go with ros2/rmw#393 latest rolling fix. otherwise, it breaks build.

[pablogs9]

Yes, sure, the point is that it works on Linux, but somehow, it is failing to initialize in embedded platforms (ESPIDF and Renesas RA by now), which we are currently investigating.

[fujitatomoya]

thanks! let me know if anything i can help on my side.

[pablogs9]

@fujitatomoya, it was our micro-ROS fork of RCL that failed to validate the enclave as being NULL. Fixed here: micro-ROS/rcl#24

I'm restoring this PR