pr: ignore empty line after newline char

[jfinkels]

Depends on #10303.

Fix a bug in pr where a form feed character (\f) immediately
following a newline character (\n) was incorrectly rendered as an
extra blank line in the output when running with the -f option. After
this commit, the empty line is correctly ignored.

[codspeed-hq]

CodSpeed Performance Report

Merging this PR will degrade performance by 15.33%

_{Comparing jfinkels:pr-newline-form-feed (92b79f5) with main (cefe8a1)¹}

Summary

⚡ 1 improved benchmark
❌ 7 regressed benchmarks
✅ 274 untouched benchmarks
⏩ 38 skipped benchmarks²

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Memory	sort_long_line[160000]	858.6 KB	716.6 KB	+19.81%
❌	Memory	sort_unique_locale[500000]	33 MB	36.9 MB	-10.58%
❌	Memory	sort_accented_data[500000]	21.6 MB	25.5 MB	-15.33%
❌	Memory	sort_key_field[500000]	29.2 MB	32.8 MB	-11.14%
❌	Memory	sort_numeric[500000]	46 MB	48.7 MB	-5.5%
❌	Memory	sort_ascii_only[500000]	21.9 MB	25.5 MB	-14.35%
❌	Memory	sort_mixed_data[500000]	22.4 MB	25.8 MB	-13.2%
❌	Memory	du_wide_tree[(5000, 500)]	1.2 MB	1.2 MB	-3.92%

Footnotes

1. No successful run was found on main (5973613) during the generation of this report, so cefe8a1 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

2. 38 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tail/follow-name (fails in this run but passes in the 'main' branch)

[ChrisDryden]

This overall made the GNU test failures go down by 10 but also introduced 6 regressions, Its due to issues in the line_num count. I'm trying to figure out locally what the missing conditions are.

EDIT: My bad was not comparing the right upstream main, there were 6 test fixes and no regressions

[ChrisDryden]

I've been messing around for a while to see what are the causes of the regressions before the most recent set of PR changes:

     9Pr
     9Pr.p
     9Pr.r
     9Pt
     9Pt.p
     9Pt.r

These ones began failing after: "597361324 (pr: ignore empty line after form feed char #10331)" but I m unable to replicate what the fix would be for these since there are multiple issues that appear to cancel out that make these tests pass. So in that other PR, even though the logic was correct it still caused these regressions