Write vel(2) to 2D probe files and drop the per-step stdout print

[Mohit-Ak]

For a plain 2D case (no bubbles, no hypoelasticity), s_write_probe_files wrote only (t, rho, vel(1), pres) to D/probe<i>_prim.dat and then printed time = ... rho = ... pres = ... to stdout on every step. The 3D branch and the 2D bubble/hypoelastic branches all include vel(2), so the plain-2D branch was just incomplete, and the print looks like leftover debugging — one line per probe per step in the job log.

This adds vel(2) to that branch and drops the print. The write is wrapped in the same #:if not MFC_CASE_OPTIMIZATION or num_dims > 1 guard the neighbouring bubble and hypoelastic branches use, since vel(2) does not exist under 1D case optimization. The 2D probe file now has 5 columns, (t, rho, vel(1), vel(2), pres).

Testing. There was no test covering the plain-2D probe path (the only 2D probe case is the hypoelastic one, which already writes vel(2)), so this adds 2D -> 1 Fluid(s) -> probe (tests/863E4401) with a probe at (0.5, 0.5) and a non-zero vel(2) on patch 2 so the new column is actually discriminating rather than a column of zeros. Generated on pristine master the golden shows 4 values per step; running the fixed binary against that golden fails with Variable count didn't match for D/probe1_prim.dat, and the committed golden (regenerated with the fix) has 5 values per step with the fourth being the 0.1 y-velocity.

$ ./mfc.sh test -o 863E4401 18DB27D5 --no-build      # new test + existing 2D hypoelastic probe test
2 passed
$ grep -c 'time =' <simulation stdout for tests/863E4401/case.py>
0
$ ./mfc.sh lint         # 669 passed
$ ./mfc.sh precheck -j 4    # 7/7 OK

Any external 2D golden probe files (outside this repo) will need regenerating for the extra column, as noted in the issue.

Fixes #1841

[github-actions]

Lines of Code

File	Lines	Diff
src/simulation/m_data_output.fpp	1403	+1

Directory	Lines	Diff
simulation	26387	+1
total	44075	+1