Fix l1 norm eval

[Luke-Pratley]

Fixing an issue with template matching and eigen when evalulating sopt::l1_norm.

Also ensuring that the operator norms can be passed to the algos directly

[codecov-io]

Codecov Report

Merging #254 into development will increase coverage by 0.08%.
The diff coverage is 94.11%.

@@               Coverage Diff               @@
##           development     #254      +/-   ##
===============================================
+ Coverage        95.45%   95.53%   +0.08%     
===============================================
  Files               66       66              
  Lines             3671     3670       -1     
===============================================
+ Hits              3504     3506       +2     
+ Misses             167      164       -3

Impacted Files	Coverage Δ
cpp/sopt/imaging_forward_backward.h	75% <ø> (ø)	⬆️
cpp/sopt/power_method.h	97.26% <ø> (ø)	⬆️
cpp/sopt/imaging_padmm.h	88.46% <0%> (ø)	⬆️
cpp/tests/primal_dual.cc	100% <100%> (ø)	⬆️
cpp/sopt/l1_proximal.h	95.65% <100%> (+0.03%)	⬆️
cpp/sopt/proximal.h	100% <100%> (ø)	⬆️
cpp/tests/wavelets.cc	99.52% <100%> (+0.01%)	⬆️
cpp/sopt/imaging_primal_dual.h	95.57% <100%> (+0.74%)	⬆️
cpp/sopt/padmm.h	95.83% <100%> (ø)	⬆️
cpp/sopt/forward_backward.h	89.83% <100%> (ø)	⬆️
... and 5 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 f9ab56a...655a2d8. Read the comment docs.