Add ToolPath and ToolPathSegment class

[Levi-Armstrong]

This adds two classes ToolPath and ToolPathSegment. These two classes provided additional information for traceability during filtering operations and integration into tesseract_qt rendering. @marip8 Would you have time to take a look at this PR?

[marip8]

What are you planning to use these for?

[Levi-Armstrong]

Primarily for tool path visualization in tesseract_qt and filtering operations in a larger task composer graph which removes unreachable waypoints, small segments and breaking up large segments at configuration flips.

[codecov]

Codecov Report

Merging #832 (be58dda) into master (015a9eb) will increase coverage by 0.01%.
The diff coverage is 93.10%.

❗ Current head be58dda differs from pull request most recent head ee5325f. Consider uploading reports for the commit ee5325f to get more accurate results

@@            Coverage Diff             @@
##           master     #832      +/-   ##
==========================================
+ Coverage   90.19%   90.20%   +0.01%     
==========================================
  Files         234      238       +4     
  Lines       14127    14185      +58     
==========================================
+ Hits        12742    12796      +54     
- Misses       1385     1389       +4     

Impacted Files	Coverage Δ
...ct_common/include/tesseract_common/serialization.h	89.36% <ø> (ø)
...seract_common/include/tesseract_common/tool_path.h	66.66% <66.66%> (ø)
...ommon/include/tesseract_common/tool_path_segment.h	66.66% <66.66%> (ø)
tesseract_common/src/tool_path_segment.cpp	92.85% <92.85%> (ø)
tesseract_common/src/tool_path.cpp	100.00% <100.00%> (ø)

[marip8]

Primarily for tool path visualization in tesseract_qt and filtering operations in a larger task composer graph which removes unreachable waypoints, small segments and breaking up large segments at configuration flips.

Sounds familiar...

Why not just use a composite instruction of Cartesian waypoints or move instructions instead? Seems like that probably has all the information necessary to do what you're looking for without having to add and maintain a new set of classes

[Levi-Armstrong]

Why not just use a composite instruction of Cartesian waypoints or move instructions instead? Seems like that probably has all the information necessary to do what you're looking for without having to add and maintain a new set of classes

Yea, I thought about that but was not sure. It would be good to meet and discuss.