Document why the factor 1/3 is correct, even in 2d

[bangerth]

@MFraters asked a great question on the mailing list, namely why the compressible strain has the factor 1/3 in eps(u)-1/3(div u)I, even for 2d computations.

I think we ultimately all agreed that the 1/3 really is correct, because we think of 2d flow as a cross section of an infinitely long 3d domain in the z-direction. In other words, it all comes down to what exactly we mean when we say that we do computations in 2d. But, since the question is valid and the answer not obvious, it would be worthwhile documenting this in the manual.