PR ended pretty big

PR ended pretty big

Not really? The majority is the auto-generated attribute tables for primitives in the docs. Kind of unavoidable when material properties change.

The actual changes are minimal: npot out and mag/minFilter in:

 docs/components/material.md                       |  3 ++-
 src/components/layer.js                           |  2 +-
 src/components/material.js                        |  6 +++++-
 src/extras/primitives/primitives/a-sky.js         |  2 +-
 src/extras/primitives/primitives/a-videosphere.js |  2 +-
 src/utils/material.js                             | 23 +++++++++++------------
 6 files changed, 21 insertions(+), 17 deletions(-)

I verified with some console.log that by default on a texture we have

wrapS = THREE.ClampToEdgeWrapping;
wrapT = THREE.ClampToEdgeWrapping;
magFilter = THREE.LinearFilter;
minFilter = THREE.LinearMipmapLinearFilter;

so here the default on the minFilter and magFilter properties in this PR matches that and just setting minFilter to linear indeed does the same thing than the previous code.

Also I confirm this is still needed to set minFilter here.
I don't see any difference between minFilter LinearMipmapLinearFilter and LinearFilter on a 4096x2048 image on Quest 3
https://github.com/aframevr/aframe/blob/master/examples/boilerplate/panorama/puydesancy.jpg
but I see a difference in sharpness on a 8192x4096 image with a bit of flickering at times
https://github.com/c-frame/aframe-stereo-component/blob/master/examples/basic_image/textures/L.jpg
Of course it's even better with a mono equirect layer and there is no flickering at all.

So the changes in the PR looks good to me. Thanks for the work @mrxz
If some third party components used npot: true, it will need to be updated to use minFilter: linear instead.

Thanks!