Use specified threshold instead of DEFAULT_CHUNK_THRESHOLD in pickchunksize

[chriselrod]

No description provided.

[KristofferC]

Wow..

[chriselrod]

Wow..

Git blame says it's been like this for 4 years!

[chriselrod]

Is this test always so unreliable?

Error During Test at /home/runner/work/ForwardDiff.jl/ForwardDiff.jl/test/DualTest.jl:400
  Test threw exception
  Expression: dual_isapprox(NESTED_FDNUM ^ PRIMAL, exp(PRIMAL * log(NESTED_FDNUM)))
  DomainError with -8.830391984918102e18:

[codecov-commenter]

Codecov Report

Merging #552 (f483f2e) into master (9ec5e6b) will increase coverage by 0.02%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #552      +/-   ##
==========================================
+ Coverage   85.10%   85.13%   +0.02%     
==========================================
  Files           9        9              
  Lines         826      841      +15     
==========================================
+ Hits          703      716      +13     
- Misses        123      125       +2     

Impacted Files	Coverage Δ
src/prelude.jl	36.36% <100.00%> (-1.14%)	⬇️
src/apiutils.jl	100.00% <0.00%> (ø)
src/jacobian.jl	99.29% <0.00%> (+0.01%)	⬆️
src/gradient.jl	98.90% <0.00%> (+0.03%)	⬆️
src/partials.jl	84.34% <0.00%> (+0.13%)	⬆️
src/dual.jl	73.51% <0.00%> (+0.20%)	⬆️
src/config.jl	87.27% <0.00%> (+0.23%)	⬆️

---

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 9ec5e6b...f483f2e. Read the comment docs.

[KristofferC]

Yes, they are.