Fix numpy>=2 compatibility in cdoTimeAverager and frepytoolsTimeAverager

[Copilot]

Describe your changes

NumPy 2.0 changed type promotion rules (NEP 50), causing both cdoTimeAverager and frepytoolsTimeAverager to produce different numerical results than with numpy 1.x. The issue stems from implicit type conversions during weight calculations for time-averaged data.

Changes to cdoTimeAverager.py:

• Explicitly cast time_bnds arrays to float64 after reading from NetCDF to bypass NumPy 2.0's changed promotion behavior
• Replace Python's sum() with np.sum(dtype=np.float64) for consistent accumulation across numpy versions
• Optimize by storing transposed array instead of calling np.moveaxis() twice
• Remove redundant [:].copy() operations since subtraction already creates new arrays

Changes to frepytoolsTimeAverager.py:

• Explicitly cast time_bnds arrays to float64 to ensure consistent precision
• Replace Python's sum() with numpy.sum(dtype=numpy.float64) in weight calculation
• Optimize moveaxis operation to avoid redundant computation
• Replace Python's sum() with numpy.sum(dtype=numpy.float64) in both weighted and unweighted calculation loops

# Before (implicit type promotion, potential inconsistency)
wgts = (np.moveaxis(time_bnds, 0, -1)[1][:].copy() - 
        np.moveaxis(time_bnds, 0, -1)[0][:].copy())
wgts_sum = sum(wgts)

# After (explicit float64, consistent across numpy versions)
time_bnds = np.asarray(time_bnds, dtype=np.float64)
time_bnds_transposed = np.moveaxis(time_bnds, 0, -1)
wgts = time_bnds_transposed[1] - time_bnds_transposed[0]
wgts_sum = np.sum(wgts, dtype=np.float64)

Impact:
The fix ensures both CDO and frepytools produce consistent, identical results across numpy 1.x and 2.x. The critical test test_compare_unwgt_fre_cli_to_unwgt_cdo now passes with 0.0 differences (previously had 37,110 non-zero differences at ~1e-10 level).

Issue ticket number and link (if applicable)

Addresses #711

Checklist before requesting a review

• I ran my code
• I tried to make my code readable
• I tried to comment my code
• I wrote a new test, if applicable
• I wrote new instructions/documentation, if applicable
• I ran pytest and inspected it's output
• I ran pylint and attempted to implement some of it's feedback
• No print statements; all user-facing info uses logging module

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[ilaflott]

@copilot

• i think your approach is the right idea, but the pipeline is not succeeding, so maybe we are looking in the wrong place
• maybe the frepytools answer changed, chase down this possibility
• start your work by creating a fre-cli environment

[Copilot]

@copilot

• i think your approach is the right idea, but the pipeline is not succeeding, so maybe we are looking in the wrong place
• maybe the frepytools answer changed, chase down this possibility...

You were right! The frepytools answer was indeed changing with numpy>=2 due to the same type promotion issues.

I fixed frepytoolsTimeAverager.py by:

• Adding explicit float64 casting for time_bnds arrays
• Replacing Python's sum() with numpy.sum(dtype=numpy.float64)
• Optimizing the moveaxis operation
• Fixing sum operations in both weighted and unweighted calculation loops

The test test_compare_unwgt_fre_cli_to_unwgt_cdo now passes - all differences are now 0.0 instead of the 1e-10 errors we were seeing.

Fixed in commit 90f7803.

[codecov]

Codecov Report

❌ Patch coverage is 90.90909% with 1 line in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (use-numpy-v2@d9e0e9f). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
...p/generate_time_averages/frepytoolsTimeAverager.py	85.71%	1 Missing ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##             use-numpy-v2     #721   +/-   ##
===============================================
  Coverage                ?   82.77%           
===============================================
  Files                   ?       68           
  Lines                   ?     4529           
  Branches                ?        0           
===============================================
  Hits                    ?     3749           
  Misses                  ?      780           
  Partials                ?        0           

Flag	Coverage Δ
unittests	82.77% <90.90%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
fre/app/generate_time_averages/cdoTimeAverager.py	100.00% <100.00%> (ø)
...p/generate_time_averages/frepytoolsTimeAverager.py	80.29% <85.71%> (ø)

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d9e0e9f...4ea8fa3. Read the comment docs.

🚀 New features to boost your workflow:

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

[ilaflott]

looks backward compatible with numpy=1.26.4:

(fre-cli) ldt-4788479: fre-cli $] pytest fre/app/generate_time_averages/tests
================================================== test session starts ===================================================
platform linux -- Python 3.12.12, pytest-9.0.2, pluggy-1.6.0
rootdir: /home/Ian.Laflotte/Working/fre-cli
configfile: pyproject.toml
plugins: cov-7.0.0, zarr-3.1.5
collected 35 items                                                                                                       

fre/app/generate_time_averages/tests/test_cdoTimeAverager.py .                                             [  2%]
fre/app/generate_time_averages/tests/test_combine.py ...                                                   [ 11%]
fre/app/generate_time_averages/tests/test_generate_time_averages.py ..............xx.x.....                [ 77%]
fre/app/generate_time_averages/tests/test_wrapper.py xxx...x.                                              [100%]
...
====================================== 28 passed, 7 xfailed, 27 warnings in 50.58s =======================================
(fre-cli) ldt-4788479: fre-cli $] python -i -c 'import numpy; print(numpy.__version__);'
1.26.4