Bump changelogs and release 2.8.1

[luca-della-vedova]

Will retag and bloom-release to rolling afterwards

[mxgrey]

I was hoping to get #332 before the next release. If it's too late then I can just follow up with another patch release later.

[aaronchongth]

The sync would be 17th Jan (Saturday) SGT 4pm, assuming we want the builds to settle with 24 hours to spare (this duration is a wild guess, it might need longer), we will need to revert this bump, get #332 merged in, and bump anew before Friday EOD. Should be possible with #332 being a small change

[luca-della-vedova]

True but we are technically in freeze for the sync and only fixes that fix regressions will be considered, I'm not actually sure this will be released before the freeze since it doesn't technically fix a build regression, although the package is kinda useless without it

[mxgrey]

I'm noticing that CI is failing now for rmf_demos, related to the Qt5 / Qt6 issues:

  CMake Error at /opt/ros/rolling/share/rviz_common/cmake/ament_cmake_export_dependencies-extras.cmake:21 (find_package):
    Found package configuration file:
  
      /usr/lib/x86_64-linux-gnu/cmake/Qt5/Qt5Config.cmake
  
    but it set Qt5_FOUND to FALSE so package "Qt5" is considered to be NOT
    FOUND.  Reason given by package:
  
    The Qt5 package requires at least one component
  
  Call Stack (most recent call first):
    /opt/ros/rolling/share/rviz_common/cmake/rviz_commonConfig.cmake:41 (include)
    CMakeLists.txt:30 (find_package)
  
  
  -- Configuring incomplete, errors occurred!

I can't think of a reason that Qt5 would be left out of the CI since we run rosdep before building.

[luca-della-vedova]

I think this is just another manifestation of ros2/rviz#1635 (review). It should be fixed once the rolling sync is done in a few days.
We could have gone for a "green CI" way and only fix things when we could have got green CI but that would have forced us to wait for the next sync to have the rviz fix in, but rmf is failing already and would have been removed from Rolling in the meanwhile.
Instead, we tested in the testing ROS repos to use the binaries before they make it to the public and we validated that everything works, so this red CI is expected but a fluke.
And to be honest Rolling has been broken for a long time so this failure is really nothing new, it's just that it wasn't caught by CI because we haven't been running CI for some time.
Looking at the logs of the builds in the buildfarm, you can see how it's been red for a long time, and just only became green because we merged and released the fix in rmf_visualization.