Separate the visibility in the editor from the visibility in the game for instances, layers and effects

[Bouh]

This PR adds separate render and editor visibility, allowing instances to be hidden in the editor without affecting rendering.
Becasue clicking an instance from the instance list focus on it in the 3D, I changed it to no longer moves the camera when we hide the instance in editor. Hover an instance in the list, press F to focus in the editor & if the instance isn't visible in the isntance list F by hovering the panel scroll to the element. Except on mobile where clicking still focuses the instance in 3D (no change)
The instance list also gets improved panel headers now have icons for better consistency.

Compatibility is preserved for previous project, the hidden instances and layers that were hidden are setup hidden in the render view not in editor.

[github-actions]

Storybook for this branch: latest · commit 23ea0e9

_{Updated automatically after each Storybook build.}

[github-actions]

⏱️ Game engine benchmarks: ✅ no significant change on the 31 benchmarks

All the 31 benchmarks and how they are measured

	Benchmark	Base	This version	Change	Range between rounds
≈	platformer objects jumping in loop(1000x)	126.9ms	131.3ms	+3.4%	-9.7% to +17.3%
≈	hitBoxesCollisionTest of 10 objects with 10 objects(10000x)	82.7ms	81.8ms	-1.0%	-2.6% to +1.8%
≈	hitBoxesCollisionTest of 30 objects with 30 objects(1112x)	24.8ms	25.1ms	+1.1%	-5.5% to +10.8%
≈	hitBoxesCollisionTest of 100 objects with 100 objects(100x)	8.340ms	8.337ms	-0.0%	-2.3% to +12.0%
≈	hitBoxesCollisionTest of 500 objects with 500 objects(4x)	1.880ms	1.888ms	+0.4%	-5.4% to +2.9%
≈	hitBoxesCollisionTest of 1000 objects with 1000 objects(1x)	1.039ms	1.039ms	+0.0%	-6.2% to +6.1%
≈	hitBoxesCollisionTest of 1 object with 1000 objects(1000x)	53.5ms	53.6ms	+0.1%	-5.1% to +5.2%
≈	hitBoxesCollisionTest of 10 objects with 1000 objects(100x)	11.0ms	11.0ms	-0.1%	-4.7% to +4.3%
≈	hitBoxesCollisionTest of 1000 objects with 1 object(1000x)	57.7ms	56.3ms	-2.5%	-5.7% to +2.4%
≈	hitBoxesCollisionTest of 500 objects with themselves(4x)	1.305ms	1.305ms	+0.0%	-2.1% to +1.3%
≈	setAngle and setLength(100000x)	11.4ms	11.4ms	+0.1%	-0.7% to +0.6%
≈	convertCoords(100000x)	21.4ms	21.4ms	+0.0%	-0.5% to +0.6%
≈	convertInverseCoords(100000x)	21.4ms	21.4ms	+0.1%	-0.2% to +1.1%
≈	collisionTest between two overlapping rectangles(120000x)	47.8ms	48.2ms	+0.8%	-4.5% to +4.8%
≈	collisionTest between two non overlapping rectangles(120000x)	11.1ms	11.2ms	+0.3%	-8.1% to +3.1%
≈	getAABB of a non rotated, default center(180000x)	19.1ms	19.0ms	-0.8%	-5.9% to +0.5%
≈	getAABB of a rotated, default center(180000x)	19.1ms	19.0ms	-0.7%	-5.9% to +0.5%
≈	getAABB of a non rotated, non default center(180000x)	20.1ms	20.1ms	-0.1%	-5.1% to +2.6%
≈	getAABB of a rotated, non default center(180000x)	61.0ms	59.8ms	-1.9%	-5.7% to +6.9%
≈	getAABB of a non rotated sprite, with custom hitboxes, origin and center(60000x)	19.6ms	19.7ms	+0.4%	-1.3% to +1.5%
≈	getAABB of a rotated sprite, with custom hitboxes, origin and center(60000x)	67.4ms	67.9ms	+0.7%	-0.1% to +1.4%
≈	get(400000x)	11.6ms	11.6ms	-0.0%	-0.2% to +0.1%
≈	Scene: Static sprites (10000)	5.228ms	5.123ms	-2.0%	-11.4% to +7.4%
≈	Scene: Moving and rotating sprites (5000)	7.059ms	7.251ms	+2.7%	-12.6% to +10.9%
≈	Scene: Collisions and separation (500 vs 500)	16.4ms	15.3ms	-6.5%	-8.2% to +3.8%
≈	Scene: Physics 2D (500 bodies)	3.529ms	3.416ms	-3.2%	-9.6% to +3.0%
≈	Scene: Physics 3D (500 bodies)	13.3ms	13.3ms	-0.5%	-1.9% to +3.6%
≈	Scene: Platformer characters (300)	1.064ms	1.039ms	-2.4%	-18.4% to +2.5%
≈	Scene: Text updates (1000)	55.7ms	55.6ms	-0.2%	-3.2% to +3.4%
≈	Scene: Objects creation and deletion (100 per frame)	2.124ms	1.899ms	-10.6%	-25.1% to +8.0%
≈	Scene: Heavy events (variables, loops, strings)	1.607ms	1.485ms	-7.6%	-14.3% to +1.9%

✅ The counters (draw calls, objects count and final state) of the 9 scene benchmarks run on both versions are unchanged.

Comparing ac2408bc to its base 1de9199b (on a pull request, the version tested is the pull request merged into its base), run alternately 4 times on the same machine (AMD EPYC 7763 64-Core Processor, 4 cores, rendering without GPU). Times are the median duration of a batch of iterations (micro benchmarks) or of a frame (scene benchmarks). A change is reported only when larger than 5% and than the differences between the rounds of a same version, and measured by every pair of rounds.