fix: preserve precision when scaling binary tables

[jamesrafe]

No description provided.

[codecov]

Codecov Report

❌ Patch coverage is 80.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 47.37%. Comparing base (870e4ee) to head (0e5a427).

Files with missing lines	Patch %	Lines
src/hdu/bintable.jl	80.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #48      +/-   ##
==========================================
+ Coverage   46.35%   47.37%   +1.01%     
==========================================
  Files          17       17              
  Lines        2550     2550              
==========================================
+ Hits         1182     1208      +26     
+ Misses       1368     1342      -26     

☔ View full report in Codecov by Harness.
📢 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.

[barrettp]

1. Instead of an array of Tuples, the data can be a Tuple of arrays, which is simpler,
2. The test needs to ensure that the element type of the array is the smallest useable type, e.g., Float32, not just Float64. Arrays can be very large, so the returned array type needs to be the appropriate type. It is not clear that muladd() ensures the best type is returned.

[jamesrafe]

Hi @barrettp , I changed the fixture to a tuple of column arrays and added explicit eltype checks. The tests now verify that results remain Float32 when sufficient, while using Float64 for large TZERO.

[barrettp]

PR #50 likely supersedes this patch. Without a description of the problem, it cannot be directly determined. Closing for now.