🐛 Fix TIFFWSIReader read_bound

[vqdang]

Emergency bugfix per @John-P request.
The culprit is reading bound doesn't use the adjusted bounds as have been done in OpenSlideReader.

[shaneahmed]

Thanks @vqdang and @John-P Please can you add a simple test to make sure this produces expected results.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.86%. Comparing base (621a857) to head (a2c8afe).
Report is 1 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #777      +/-   ##
===========================================
- Coverage    99.89%   99.86%   -0.04%     
===========================================
  Files           69       69              
  Lines         8650     8650              
  Branches      1653     1654       +1     
===========================================
- Hits          8641     8638       -3     
- Misses           1        4       +3     
  Partials         8        8              

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

[shaneahmed]

Please can you also check if it is related to #499 ?

[Abdol]

I have added a simple test but not completely sure if it fully tests for the fixed bug. @vqdang @John-P I'd appreciate it if you can give the test a review.

[Abdol]

A kind follow up @vqdang @John-P. I'd appreciate it if you can give the test a review. Thanks!

[measty]

I think this also needs a fix along similar lines for read_rect

[Abdol]

A kind follow up @vqdang @John-P. I'd appreciate it if you can give the test a review. Thank you!

[measty]

I've added a few changes to this PR too.

• fix for read_rect along similar lines to read_bounds (it was bugged in the same way)
• a few changes to make the reader compatible with a wider variety of tiffs
• a change to metadata in wsireader that fixes some performance issues on some wsis

[Abdol]

After updating the sample OME TIFF, the bug can be reproduced using the existing level consistency tests. So, I have removed the newly-added test. This PR should be good to go now.

[shaneahmed]

Thanks @Abdol , @measty and @vqdang for your help in fixing this bug.