Integrate automated tablet placement

[hello-amal]

Description

This PR integrates the automated behavior of displaying a tablet to a user into the web app.

Testing procedure

• Pull the stretch_show_tablet repo, and pull this branch.
• Change the tool to a tablet and run stretch_configure_tool.py.
• Re-build your workspace.
• ./launch_interface.sh
• Toggle on "Show Pose", verify that it detects you when you are in Realsense camera frame and not otherwise.
• Move to different poses (stretching, sitting, standing, legs crossed), and verify it detects you.
• Toggle on realsense depth sensing while "Show Pose" is on, and verify both are overlaid on.
• Verify that the "Show tablet" button appears when "Show Poses" is checked, and dissapears otherwise.
• Uncheck "show poses". Verify the stretch_deep_perception node stops running (e.g., ros2 topic echo /body_landmarks/landmarks_3d and verify nothing publishes).
• Drive the robot near you. Check "Show Poses." Click "Show tablet." Verify it works.
• Repeat the above from several different base positions/orientations relative to you, and several different body postures.
• Run "Show tablet" and then cancel it, verify that works.
• Repeat the above, canceling at each distinct joint motion.
• Run "Show tablet" and then command a joint motion, verify it cancels.
• Repeat the above at every distinct joint motion in "Show Tablet."
• Re-run the performance benchmarking documented here, verify that this code doesn't hinder performance.

Before opening a pull request

From the top-level of this repository, run:

• pre-commit run --all-files

To merge

• Squash & Merge

[hello-amal]

Tested on 3053. While I didn't run all the above tests on 3053, across multiple robots the above tests have been run. Merging it in due to time constraints.