feat(ansible): set default value for install_dev variable

[mitsudome-r]

Description

Due to the refactoring, it seems like the evaluator for testing scenario testing is failing. It seems like it is due to change in expected variables for universe playbook, where the refactored version expects install_devel to be set.

I modified the playbook to set true as default. However, the modification now prompts the user to install dev_tools. If there are any other way to set default value for the variable without prompting, please let me know.

Tests performed

Tested locally with ./setup-dev-env.sh, and tested with the evaluator.

Effects on system behavior

The new script now prompts the user to install dev_tools when the script is executed.

Not applicable.

Pre-review checklist for the PR author

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

• [x] I've confirmed the contribution guidelines.
• [x] 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.

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.

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

[oguzkaganozt]

I think scenario testing might not be working due to missing cuda-command-line-tools when installing only runtime bindings for CUDA and TensorRT, could you try adding it and try again ? And It's also a good idea to have a prompt for installing development dependencies. Thanks @mitsudome-r.

[oguzkaganozt]

In my opionion, the name of "install_dev" needs more refactoring.

For the time being, "install_dev" is the flag for installing dev_tools (git lfs, pre-commit, clang-format and so on), and also nvidia cudnn/tensorrt.

The problem is that git-lfs, pre-commit are indeed development tools that is not required in run time, while nvidia cudnn / tensorrt is required for full usage of autoware.

We need to divide the usage of this variable before we add more modifications on its pre-defined value

Actually right now it is installing only runtime bindings both for CUDA and tensorRT but not dev packages and header files, so cuda and tensorRT dependent packages should work without any problem. Last time I tested I couldn't see any problems related to these packages. If you have any related issues could you please share ? Thanks for the feedback, @Owen-Liuyuxuan