Optimize 2D rho grid memory use

[dafeda]

Use NumPy broadcasting in calc_rho_for_2d_grid_layer instead of
flattened mesh arrays. This preserves the result while avoiding large
intermediate allocations.

• PR title captures the intent of the changes, and is fitting for release notes.
• Added appropriate release note label
• Commit history is consistent and clean, in line with the contribution guidelines.
• Make sure unit tests pass locally after every commit (git rebase -i main --exec 'just rapid-tests')

When applicable

• When there are user facing changes: Updated documentation
• New behavior or changes to existing untested code: Ensured that unit tests are added (See Ground Rules).
• Large PR: Prepare changes in small commits for more convenient review
• Bug fix: Add regression test for the bug
• Bug fix: Add backport label to latest release (format: 'backport release-branch-name')

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.03%. Comparing base (5d35e74) to head (8e899c1).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #13425      +/-   ##
==========================================
- Coverage   90.03%   90.03%   -0.01%     
==========================================
  Files         459      459              
  Lines       31992    31992              
==========================================
- Hits        28804    28803       -1     
- Misses       3188     3189       +1     

Flag	Coverage Δ
cli-tests	37.49% <100.00%> (ø)
fuzz	44.24% <0.00%> (ø)
gui-tests	66.94% <0.00%> (+0.02%)	⬆️
performance-and-unit-tests	78.05% <100.00%> (-0.01%)	⬇️
test	45.86% <0.00%> (ø)

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

Files with missing lines	Coverage Δ
src/ert/field_utils/field_utils.py	93.56% <100.00%> (ø)

... and 2 files with indirect coverage changes

[codspeed-hq]

Merging this PR will not alter performance

✅ 36 untouched benchmarks

---

_{Comparing dafeda:calc-rho-memory (8e899c1) with main (5d35e74)}

[xjules]

Wondering whether this shift (+0.5) is necessary at all? Since thus we need to reflect this in the observations too (ie. like in heat equation).
I guess the increment can vary and thus it makes sense.....

[larsevj]

Could you rebase to ensure the snapshot test passes?