Unexpected Convergence Behavior in plateWithHole_linElast_2D.cpp

[Bahari95]

I tested this code and added the L2 error for σxx​. However, the convergence order with respect to refinement appears unstable, producing unexpected results. This issue is also evident in the stress values themselves.

https://github.com/gismo/gsElasticity/blob/master/examples/plateWithHole_linElast_2D.cpp

XX-stress at the top of the circle: 3.027940e+00 (computed), 3.000000e+00 (analytical) DoFs = 2244
XX-stress at the top of the circle: 2.659625e+00 (computed), 3.000000e+00 (analytical) DoFs = 8580
XX-stress at the top of the circle: 2.994378e+00 (computed), 3.000000e+00 (analytical) DoFs = 33540