Test world scn update

[jmbau]

Plots plot03.3d, plot 04.3d and plot05.3d are intended to be didactic:

• shows that a square texture applied on a square height field can generate a rectangular terrain (plot03.3d)
• shows that a terrain plot can be vertical translated (plot04.3d)
• shows that a full black color used in a rgb (not rgba) texture is fully transparent
• shows that a partially 1% transparent rgba color (plot04.tex, lava lake area) is detected as a liquid surface

With these examples, I discovered that:

• terrain textures seems to be repeated from one border to the other one: the color of the "starting" border is slightly visible on the "ending" border (and/or the color of the "ending" border is slightly visible on the "starting" border, I don't know).
• I'm not sure about that, but it seems that height field always finishes horizontally at their lowest Y positions, as if lowest Y height (z value) were repeated at lowest Y-1 position:
  
  Maybe that a 128x128 pixels height field produces a 129x129 points grid ???
• all texture area pixels which are detected as transparent (i.e. full black colored pixel in a rgb texture and any pixel with at least 1% of transparency in a rgba texture) are fully transparent in map view. For my mind it's not an issue, it's just an observation.

[hsanjuan]

@jmbau thanks for your observations... if anything is worth fixing later let's open a specific issue about it.

I have rebased/merged your open PRs now and you can branch off from master and should have all your changes. Thanks!