469 task rewrite thruster interface

[albertomors]

No description provided.

[Andeshog]

Remove the python package and change the name of the new cpp package back to the original name

[jorgenfj]

Also update the thrust_update_rate: 10.0 # [Hz] in orca.yaml
Should be higher than 10.0

[chrstrom]

First pass of the C++ stuff.

As a general note, you might want to remove most of the inline comments. As a general rule of thumb, you should have a very good reason for adding inline codes (if you can't avoid some bit magic or some specific equation etc), as the code itself should do the explaining: Comments may lie (and/or become outdated), but the code wont!

This can most notably be seen in the suggestions to use std::algorithm, as that makes things a lot more explicit (especially when defining the lambdas outside the algorithm call and giving it a fitting name!

Also note that the suggestions here are free-balled, so make sure everything compiles and works as expected 😅

[Andeshog]

Almost there now. As mention in one of the comments I think it will be a good idea to remove the timer callback and just send pwm in the thruster_forces subscriber callback instead. The cpp part looks good enough for me, but I will let @chrstrom be the final judge for that.

[jorgenfj]

I unresolved the comment regarding the debug publisher from my previous review. The publisher should only publish when we actually want to debug.

Also I had to include:
#include
#include

in order to compile pwm_to_i2c_data

I got this error squiggle in vscode:
function "ThrusterInterfaceAUVDriver::pwm_to_i2c_data" returns incomplete type "std::array<uint8_t, 2UL>" (aka "std::array<unsigned char, 2UL>")C/C++(409)

Although seems like the automated build process didn't encounter this so not sure if it should be a concern

[Andeshog]

aim to test the code on Orca on Friday or Sunday and then merge into develop after

bad