Use actual position when limiting desired position

[saikishor]

While integrating the joint limits into the resource manager, I discovered that the current joint_limits implementation doesn't consider the actual position when limiting the desired position. This was the case as in ROS 1 too it doesn't use it, but uses the previous command value. I think it would make sense to consider the actual position as well in this scenario

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.32%. Comparing base (f4c9dd0) to head (bf41ee1).
Report is 3 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1988      +/-   ##
==========================================
- Coverage   89.32%   89.32%   -0.01%     
==========================================
  Files         130      130              
  Lines       14522    14568      +46     
  Branches     1258     1259       +1     
==========================================
+ Hits        12972    13013      +41     
- Misses       1083     1086       +3     
- Partials      467      469       +2     

Flag	Coverage Δ
unittests	89.32% <100.00%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...s/include/joint_limits/joint_limiter_interface.hpp	50.79% <ø> (ø)
.../include/joint_limits/joint_saturation_limiter.hpp	100.00% <ø> (ø)
...limits/include/joint_limits/joint_soft_limiter.hpp	92.85% <ø> (ø)
joint_limits/src/joint_limits_helpers.cpp	91.54% <100.00%> (+0.12%)	⬆️
joint_limits/src/joint_range_limiter.cpp	84.61% <100.00%> (+0.24%)	⬆️
joint_limits/src/joint_saturation_limiter.cpp	85.29% <100.00%> (-0.34%)	⬇️
joint_limits/src/joint_soft_limiter.cpp	88.46% <100.00%> (+0.86%)	⬆️
joint_limits/test/test_joint_range_limiter.cpp	100.00% <100.00%> (ø)
joint_limits/test/test_joint_soft_limiter.cpp	99.64% <100.00%> (+0.01%)	⬆️

... and 3 files with indirect coverage changes

[christophfroehlich]

I agree, this makes sense

[saikishor]

Resolved conflicts after #1981

[saikishor]

@bmagyar It would be great to have this PR before releasing ros2_control. That way we will have the final API of joint_limits in the same tag. If that's not a problem at all, anything is fine