Move to custom task parallelism for better generalization

[mloubout]

• Binary reduction
• Generalizd task parallelization

[codecov[bot]]

Codecov Report

Merging #71 (f4df1d2) into master (863149b) will increase coverage by 0.48%. The diff coverage is 83.20%.

@@            Coverage Diff             @@
##           master      #71      +/-   ##
==========================================
+ Coverage   83.97%   84.46%   +0.48%     
==========================================
  Files          35       30       -5     
  Lines        2303     2336      +33     
==========================================
+ Hits         1934     1973      +39     
+ Misses        369      363       -6     

Impacted Files	Coverage Δ
src/TimeModeling/Modeling/fwi_objective_serial.jl	88.46% <0.00%> (-3.54%)	:arrow_down:
src/TimeModeling/Modeling/twri_objective_serial.jl	0.00% <0.00%> (ø)
src/TimeModeling/TimeModeling.jl	100.00% <ø> (ø)
...rc/TimeModeling/Modeling/lsrtm_objective_serial.jl	79.31% <40.00%> (-8.69%)	:arrow_down:
src/JUDI.jl	85.71% <50.00%> (-14.29%)	:arrow_down:
src/TimeModeling/Modeling/python_interface.jl	92.38% <87.50%> (+4.38%)	:arrow_up:
src/TimeModeling/Types/judiWeights.jl	86.84% <87.50%> (-0.24%)	:arrow_down:
src/TimeModeling/Modeling/distributed.jl	97.36% <97.36%> (ø)
...eling/Modeling/extended_source_interface_serial.jl	76.47% <100.00%> (ø)
src/TimeModeling/Types/GeometryStructure.jl	94.23% <100.00%> (+0.19%)	:arrow_up:
... and 4 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 863149b...f4df1d2. Read the comment docs.

[mloubout]

@ziyiyin97

[mloubout]

Good to go?

[mloubout]

With no new comment gonna merge before the break. BTW this PR will allow much simpler and cleaner JUDI4Cloud

[ziyiyin97]

Yes I agree from your hard work on these distribute things.