Check that measuring a text and giving it that width is a fixed point

A span that sizes to its children measures one, is told a length, and
hands that length back -- so whether measurement is idempotent decides
whether the two chase each other. Nothing checked it.

It holds: a wrapping text in a `Dir::RIGHT` span, which is the wrap axis
and the span's own axis together, stays at 881.84 across six repaints
that change nothing. So the narrowing recorded against LAYOUT.md §4 is
not something text does on its own, and looking for the cause there is
looking in the wrong place.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

tests/idempotence.rs

@@ -0,0 +1,29 @@
+//! Whether measuring a widget and then giving it the length it reported is a
+//! fixed point, which is what a span that sizes to its children needs.
+
+use iris::harness::Harness;
+use iris::prelude::*;
+
+#[test]
+fn a_wrapping_text_in_a_span_settles_on_one_width() {
+    let mut h = Harness::new((900, 600));
+    let words = "the quick brown fox jumps over the lazy dog and keeps on running \
+                 until it reaches the end of a rather long line of text";
+    let t = wtext(words).size(16).wrap(true).add(&mut h.rsc);
+    let filler = rect(Color::BLUE).add(&mut h.rsc);
+    h.set_root((t, filler).span(Dir::RIGHT));
+
+    let mut widths = Vec::new();
+    for _ in 0..6 {
+        let r = h.region(&t.id()).unwrap();
+        widths.push(r.bot_right.x - r.top_left.x);
+        // Redrawing it changes nothing about the state, so nothing may move.
+        h.rsc.widgets_mut().get_dyn_mut(t.id());
+        h.frame();
+    }
+    println!("widths over six frames: {widths:?}");
+    assert!(
+        widths.windows(2).all(|w| w[0] == w[1]),
+        "a repaint that changed nothing moved it: {widths:?}"
+    );
+}