fix #2361 for real

[SimonDanisch]

We were freeing the camera observables when freeing the renderobject from a plot, which happened since e.g. was renderobject[:resolution] === scene.camera.resolution. Solution is to pass a copy of the camera observables which where passed directly.

Other attributes should be fine, since we lift + convert them.
I hope to figure out a real, general solution to cleaner clean ups in #2372.

[MakieBot]

Compile Times benchmark

Note, that these numbers may fluctuate on the CI servers, so take them with a grain of salt. All benchmark results are based on the mean time and negative percent mean faster than the base branch. Note, that GLMakie + WGLMakie run on an emulated GPU, so the runtime benchmark is much slower. Results are from running:

using_time = @ctime using Backend
# Compile time
create_time = @ctime fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @ctime Makie.colorbuffer(display(fig))
# Runtime
create_time = @benchmark fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @benchmark Makie.colorbuffer(display(fig))

	using	create	display	create	display
GLMakie	29.99s (29.86, 30.07) 0.09+-	18.80s (18.74, 18.89) 0.06+-	16.87s (16.77, 16.96) 0.07+-	16.64ms (16.45, 16.74) 0.10+-	49.28ms (48.81, 50.86) 0.73+-
master	29.65s (29.54, 29.84) 0.11+-	18.76s (18.68, 18.86) 0.08+-	16.70s (16.59, 16.78) 0.07+-	16.71ms (16.30, 16.90) 0.23+-	49.23ms (48.81, 50.18) 0.52+-
evaluation	+1.13%, 0.34s slower X (3.34d, 0.00p, 0.10std)	+0.20%, 0.04s invariant (0.55d, 0.33p, 0.07std)	+1.02%, 0.17s slower X (2.52d, 0.00p, 0.07std)	-0.45%, -0.07ms invariant (-0.41d, 0.46p, 0.17std)	+0.10%, 0.05ms invariant (0.08d, 0.89p, 0.63std)
CairoMakie	26.11s (26.02, 26.23) 0.07+-	18.99s (18.81, 19.11) 0.11+-	2.72s (2.69, 2.74) 0.03+-	17.73ms (17.54, 18.05) 0.18+-	23.94ms (23.77, 24.07) 0.10+-
master	25.99s (25.85, 26.08) 0.09+-	19.15s (19.02, 19.39) 0.13+-	2.73s (2.70, 2.75) 0.02+-	16.99ms (16.74, 17.33) 0.18+-	24.08ms (23.92, 24.17) 0.08+-
evaluation	+0.48%, 0.13s slower X (1.56d, 0.01p, 0.08std)	-0.81%, -0.15s faster ✓ (-1.24d, 0.04p, 0.12std)	-0.55%, -0.01s invariant (-0.65d, 0.25p, 0.02std)	+4.18%, 0.74ms slower X (4.09d, 0.00p, 0.18std)	-0.60%, -0.14ms faster ✓ (-1.51d, 0.02p, 0.09std)
WGLMakie	35.10s (33.55, 37.02) 1.26+-	32.62s (30.70, 34.81) 1.65+-	48.86s (46.46, 52.62) 2.40+-	36.95ms (35.91, 38.31) 0.81+-	120.87ms (105.74, 152.41) 14.78+-
master	35.44s (33.66, 38.80) 1.97+-	32.02s (30.21, 35.92) 2.01+-	48.21s (45.58, 53.63) 2.81+-	34.22ms (32.55, 35.58) 1.26+-	110.81ms (87.17, 148.82) 19.60+-
evaluation	-0.96%, -0.34s invariant (-0.20d, 0.71p, 1.62std)	+1.85%, 0.6s invariant (0.33d, 0.55p, 1.83std)	+1.33%, 0.65s invariant (0.25d, 0.65p, 2.61std)	+7.40%, 2.74ms slower❌ (2.59d, 0.00p, 1.03std)	+8.32%, 10.06ms noisy🤷‍♀️ (0.58d, 0.30p, 17.19std)