[amjjbonvin]

What does this PR do and why?

When complexes are consistent with all restraints, the report and the log
flagged a restraint as a possible false positive, and the report left out the
violations of N = all restraints (#59).

• report.html: the violations table shows every N with complexes, including
  the row for all restraints (all 0). The Z-score table highlights rows only
  when not all restraints are consistent; its title is then
  "Z-scores (above 1.0: possible false positive)", otherwise
  "Z-scores (all restraints are consistent)".
• disvis.log: "Possible false-positive restraints (z-score > 1.0)" only when
  not all restraints are consistent.
• README (z-score.out) and CHANGELOG updated.

The .out and .mrc files are unchanged.

Version 3.1.1, with the 3.1.0 changes under their own heading in CHANGELOG.md.

How was this tested?

• tests/test_report.py: two assertions on the existing test data (all 7
  restraints consistent): the N = 7 row is shown and no row is highlighted.
  They fail without the fix. All 28 unit tests pass.
• test-cases/check_test_cases.py (coarse, CPU): output files byte-identical.
• Example results regenerated: PRE5-PUP2 cpu/coarse, cpu/coarse-is-oa and
  cpu/fine (macOS) and gpu/coarse, gpu/fine (tintin, RTX 6000 Ada) re-run,
  output files identical to the references, log without false positives;
  RNA polymerase II reports rebuilt from the existing files (only the title
  changes, restraints 7 and 8 still flagged).

Not tested: the web service; runs 1856848 and 1856846 were checked from their
output files only.

AI assistance

Written with Claude Code; checked as above.

Checklist

• Tests cover the new and/or changed code
• Documentation updated if needed (README.md, INSTALLATION.md)
• Reference results in test-cases/*/example-results/ updated if the output changes

Related issues

Closes #59

Notes for reviewers

The log change has no unit test: the message is written in main(); it was
checked with the test-case runs above.

🤖 Generated with Claude Code