Replace rmw_connext_cpp with rmw_connextdds

[asorbini]

This PR replaces all references to rmw_connext_cpp with rmw_connextdds.

See rticommunity/rmw_connextdds #9 for a list of related PRs, and an overview of all the changes required to replace ros2/rmw_connext (rmw_connext_cpp) with rticommunity/rmw_connextdds in the ROS2 source tree.

[fujitatomoya]

these are still needed?

demos/pendulum_control/test/test_pendulum_demo.py.in

Lines 59 to 64 in ed0299d

	if rmw_implementation.startswith('rmw_connext'):
	# This output can be caused by a small QoS depth leading to samples being discarded.
	# Since we are optimizing for performance with a depth of 1, we can ignore it.
	filtered_prefixes.append(
	'PRESWriterHistoryDriver_completeBeAsynchPub:!make_sample_reclaimable'
	)

demos/pendulum_control/test/test_pendulum_teleop.py.in

Lines 63 to 68 in ed0299d

	if rmw_implementation.startswith('rmw_connext'):
	# This output can be caused by a small QoS depth leading to samples being discarded.
	# Since we are optimizing for performance with a depth of 1, we can ignore it.
	filtered_prefixes.append(
	'PRESWriterHistoryDriver_completeBeAsynchPub:!make_sample_reclaimable'
	)

[asorbini]

these are still needed?

@fujitatomoya I'm not sure, since I haven't tried removing them (and re-running the tests without them). I left them in to be on the safe side (since the tests are passing "as is"), but I can try removing them and seeing what happens.

@ivanpauno if you want I can push a commit to also remove references to opensplice, but maybe those would be better tracked via a different PR? If so, I can do that too while I'm at it.

EDIT: I realize now that creating a separate PR will only complicate things, so I'll just push a commit to this branch.

[clalancette]

PR build failed, but it seems to be doing that for all PRs. Given that this is just a change to the README files, I'm going to go ahead and merge.