Fix Phantom-related bugs (#498 and more)

[pvillacorta]

This PR solves the following:

• Troubles defining Float32 or empty phantoms with KomaMRI 0.9 #498
  Phantom(x=[]), Phantom{Float64}(x=[]) or Phantom{Float32}(x=[]) are no longer supported*, since the element type of the Phantom is Inferred from its input arguments, and defining x=[] (eltype Any) is not a good practice in Julia.
  Now, it is possible to define an empty Phantom this way:

  Phantom() # default eltype is Float64
  Phantom{Float64}()
  Phantom{Float32}()
  Phantom{Int64}()
  (...)

  *The example in this page is updated too.

• This bug when creating a motionlist from an action. I have also added some Motion constructors that might be useful.

• plot_phantom_map: now 2D or 1D phantoms in yz plane are displayed correctly. See this comment. Moreover, the prefix for the time slider in dynamic plots was incorrect ("x="). This has changed to be a ("t=").

[pvillacorta]

Buildkite CI is still failing because of JuliaGPU/oneAPI.jl#473

[cncastillo]

Is there any other problem besides oneAPI.jl not working on Julia 1.11? We could merge this and register a new version.

oneAPI.jl, at some point, will include fixes for the logical indexing problem, as there has been movement with using KernelAbstractions.jl as the backend for GPUArrays.jl.

[pvillacorta]

I added the commits from #495 that fixed bugs. It should be ready to merge

[cncastillo]

This PR has too many changes, based on the title I would have expected it to only change KomaMRIBase. Please separate the other changes, to register this new fix in komamribase, without the rest.

[cncastillo]

I think this PR also changes the name of a folder, which confused the diif's. please maintain the the name of the folders and files.

Also verify if all the docs changes are correct.

[pvillacorta]

I think it is ready now

[cncastillo]

I think the plotting changes came back (fixing another bug).

[pvillacorta]

I think the plotting changes came back (fixing another bug).

Oh, I didn't get to remove those changes at any point in the end. Don't you think it's a good idea to take the opportunity to fix those bugs as well?

[cncastillo]

Not the best practice, but let's change the title of the PR and list the bugs this PR is solving. So at least someone looking at it can understand.