Use == vs != on same operands to prove a simulator X bug

When claiming a simulator 4-state bug, the most convincing minimal repro is self-contradiction needing no external referee: e.g. VCS returns 0 for {4{en}}==4'b0101 but x for {4{en}}!=4'b0101 with en=1'bx. Run an adversarial fact-checker (e.g. Codex) on the evidence logs before publishing: it caught '307 files' that were really 208 files x modes, a 'dump makes it disappear' claim not yet backed by paired same-file runs (fixed by running the exact files with only +define+DUMP changed), and rerun-coverage overstatements. Pair every 'mode changes result' claim with identical-file runs differing only in that flag.