Fix compare Jacobian assembler for assembly mixin

That feature was broken for processes that use the OpenMP parallel assembly code. This MR re-enables the Jacobian comparisons at least for a single OpenMP thread.

1. Tests covering your feature were added?
2. Any new feature or behaviour change was documented?

• Add this merge request to the changelog

Sample output is:

#!/usr/bin/env python
import numpy as np
from numpy import nan


### counter: 117, t: 1000, element_id: 0 (begin)
# absolute tolerance of 5e-16 exceeded and relative tolerance of 4e-15 exceeded

counter = 117
nonlinear_iteration = 2
t = 1000
dt = 1000
element_id = 0

num_dof = 16
abs_tol = 5.0000000000000004e-16
rel_tol = 4.0000000000000003e-15

local_x = np.array([0.99999999999999989, 0.12216867657308761, 0.12216867657261236, 1, 0.34068545202370593, 0.12216867657286463, 0.34068545202301126, 1,
    101000, 100974.35897435935, 100974.35897435935, 101000, 100987.17948717964, 100974.35897435936, 100987.17948717969, 101000])
local_x_prev = np.array([0, 0, 0, 0, 0, 0, 0, 0,
    100000, 100000, 100000, 100000, 100000, 100000, 100000, 100000])

Jacobian_1 = np.array([
    [0.74354529914686618, 0.30760569800699966 ...