feat(lidar_centerpoint): accelerated preprocessing for centerpoint

[knzo25]

Description

While changing the point type to the new standard, realized that preprocessing was bottleneck in centerpoint.
Apologies to Amadeusz since he did almost the same already in transfusion, and I forgot until writing this PR but I also moved the preprocessing to cuda for gpu acceleration.

During tests in my desktop the detection time went from 16.9ms to 13.0ms, making preprocessing about ~1ms if we forcefully add syncs to measure the time.

Related links

Tests performed

• Tested using data from the taxi

Notes for reviewers

Interface changes

ROS Topic Changes

ROS Parameter Changes

Effects on system behavior

Pre-review checklist for the PR author

The PR author must check the checkboxes below when creating the PR.

• I've confirmed the contribution guidelines.
• The PR follows the pull request guidelines.

In-review checklist for the PR reviewers

The PR reviewers must check the checkboxes below before approval.

• The PR follows the pull request guidelines.
• The PR has been properly tested.
• The PR has been reviewed by the code owners.

Post-review checklist for the PR author

The PR author must check the checkboxes below before merging.

• There are no open discussions or they are tracked via tickets.
• The PR is ready for merge.

After all checkboxes are checked, anyone who has write access can merge the PR.

[knzo25]

@kminoda
This is by no means urgent, but it is always nice to reduce the perception latency.

if you do not feel comfortable reviewing this PR, Uetake-san, Hirabayashi-san, Tanaka-san, and Amadeusz are good candidates)

[kminoda]

@knzo25 Thank you! And yes, I am still catching up with this package (especially the CUDA part), so would be nice if someone else could review it as a co-reviewer 🙏

@amadeuszsz Hi, would you help me out reviewing CUDA related code?

[kminoda]

@knzo25 And also, would you execute Autoware Evaluator and verify that the basic scenarios would pass? Let me know if you don't know how to execute it.

[kminoda]

Thank you for the awesome PR 🚀

[knzo25]

@kminoda
Regarding the evaluation I will look into it as soon as I can (got other stuff to deliver first)

[amadeuszsz]

@amadeuszsz Hi, would you help me out reviewing CUDA related code?

@kminoda
Sure! I would like to make a review next week. Please let me know if it's urgent, I might change my priorities.
I guess you can't add me as a reviewer yet. I added my account to the contributors sheet, so I hope we will be able assign me soon.

[kminoda]

@amadeuszsz Thank you. Not an urgent task, so please review this when you have time.