[bug] Update mpm_lagrangian_force and fix Matrix constructor

[k-ye]

#1531 broke the example because it forgot to initialize self.grad. Adding the fix as well as updating mpm_lagrangian_force.py to use field.

Related issue = #1531

[Click here for the format server]

---

[archibate]

Thanks!

[archibate]

#1531 broke the example because it forgot to initialize self.grad.

So we don't have a test on this?

[codecov]

Codecov Report

Merging #1545 into master will increase coverage by 0.69%.
The diff coverage is 90.10%.

@@            Coverage Diff             @@
##           master    #1545      +/-   ##
==========================================
+ Coverage   86.10%   86.79%   +0.69%     
==========================================
  Files          19       19              
  Lines        3548     3681     +133     
  Branches      631      652      +21     
==========================================
+ Hits         3055     3195     +140     
+ Misses        365      353      -12     
- Partials      128      133       +5     

Impacted Files	Coverage Δ
python/taichi/lang/kernel.py	82.77% <ø> (+1.20%)	⬆️
python/taichi/lang/impl.py	89.41% <63.15%> (-1.14%)	⬇️
python/taichi/lang/meta.py	64.70% <75.00%> (+2.20%)	⬆️
python/taichi/lang/__init__.py	80.05% <78.94%> (+0.58%)	⬆️
python/taichi/lang/matrix.py	92.01% <89.02%> (+0.58%)	⬆️
python/taichi/lang/transformer.py	93.70% <95.00%> (+0.08%)	⬆️
python/taichi/lang/ops.py	93.33% <95.55%> (+0.20%)	⬆️
python/taichi/lang/snode.py	93.69% <100.00%> (+6.65%)	⬆️
python/taichi/lang/util.py	62.26% <100.00%> (-2.45%)	⬇️
... and 2 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update aaf0420...59f16be. Read the comment docs.

[yuanming-hu]

Cool, thanks!