Iterator norm in isapprox for Arrays

[jishnub]

This seems to improve performance and reduce allocations:

julia> A = rand(100,100);

julia> @btime isapprox($A, $A);
  13.305 μs (3 allocations: 78.21 KiB) # master
  8.063 μs (0 allocations: 0 bytes) # this PR

julia> A = rand(5000,5000);

julia> @btime isapprox($A, $A);
  106.003 ms (3 allocations: 190.74 MiB) # master
  43.122 ms (0 allocations: 0 bytes) # this PR

Currently, this is only specialized for Arrays and contiguous views of Arrays, as these are the cases with a significant performance improvement.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.84%. Comparing base (0116878) to head (ef309a6).
⚠️ Report is 42 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1378      +/-   ##
==========================================
+ Coverage   93.81%   93.84%   +0.02%     
==========================================
  Files          34       34              
  Lines       15717    15723       +6     
==========================================
+ Hits        14745    14755      +10     
+ Misses        972      968       -4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dkarrasch]

Cool. Does this have any effect on the runtime of the tests? There are probably zillions of isapprox tests, so they might add up notably.