Cap what the transcript draws, and let a LazySpan clip itself

Four things Iris asked for on 2026-09-08.

**A LazySpan no longer cares about masks.** It asserted that something
around it had called `.masked()` and refused to draw otherwise, which is
why a plain full-screen list -- the benchmark, any simple app -- panicked.
It cared only because it draws a row straddling an edge in full and relied
on somebody else to cut off the overhang; it clips itself to the box it
was offered now. Strictly stronger than the assert, which a mask *larger*
than the list's box satisfied while letting the overhang through anyway --
the fault it was written for. The transcript's `.masked()` wrapper goes
with it, and `Painter::is_masked` with that.

**Everything on the transcript screen is capped.** One rule in one place,
`client_core::text_cap`, mirrored as `TextCap.kt` with the same numbers so
a bench comparing the apps compares renderers rather than policies:

    a tool call's input    80 lines or 4 KiB   -> "Show all N lines"
    a tool call's output   80 lines or 4 KiB   -> (already was, in iris)
    a message             200 lines or 16 KiB  -> "Show all N lines"

The input is what the edit-card report needed: an Edit's old_string and
new_string arrive whole and are routinely the biggest text on screen.
Messages are capped in both apps, user and agent alike.

Three rules that took a screenshot to get right. A message is cut on a
block boundary, never mid-block -- cut to its own opening line a fence
renders as an empty panel, which reads as a fault rather than as a cap --
except a message that is one enormous block, which is truncated, since
dropping it would leave the row blank. A reply still streaming is never
capped. And the input's two blocks share one "Show all", while input and
output have their own.

**Compose stops wrapping raw text**, per Iris's call: a tool's leftover
input fields and its output pan sideways like the command already did.

`on_tap` and hold-the-edge move to `transcript-ui/src/tap.rs`, since a
message's "Show all" needs exactly what a tool card's tap already had.

docs/IRIS.md

@@ -12,7 +12,56 @@ things still stay out.
 An entry gives the date, what changed, why, and a short before/after where
 it helps judge the change without the session that made it. Newest first.
 
-## 2026-09-08 (newest): redrawing one widget cost O(its own primitives squared)
+## 2026-09-08 (newest): a `LazySpan` clips itself, and nothing is unbounded
+
+Four things you asked for, in one change.
+
+**A lazy span no longer cares about masks.** It used to assert that
+something around it had called `.masked()` and refuse to draw otherwise,
+which is why a plain full-screen list -- the benchmark, any simple app --
+panicked on its second line. Your question was the right one: it cared
+only because it draws a row straddling an edge *in full* (virtualisation
+decides which rows, never how much of one) and relied on somebody else to
+cut off the overhang. It clips itself to the box it was offered now, so a
+caller places it like any other widget. That is also strictly stronger
+than the assert was: a mask *larger* than the list's box satisfied
+`is_masked` and let the overhang through anyway, which is the fault the
+assert was written for. The transcript's own `.masked()` wrapper is gone
+with it.
+
+**Everything on the transcript screen is capped now.** One rule in one
+place -- `client_core::text_cap`, mirrored as `TextCap.kt` with the same
+numbers, so a bench comparing the apps compares renderers and not
+policies:
+
+    a tool call's input    80 lines or 4 KiB   -> "Show all N lines"
+    a tool call's output   80 lines or 4 KiB   -> (already was)
+    a message             200 lines or 16 KiB  -> "Show all N lines"
+
+The input is what your edit card needed: an `Edit`'s `old_string` and
+`new_string` arrive whole and are routinely the biggest text on screen.
+Messages are capped for the reason you gave -- both user and agent, in
+both apps.
+
+Three rules that took a screenshot to get right. A message is cut on a
+**block boundary**, not mid-block: cut to its own opening line a fence
+renders as an empty panel, which reads as a fault rather than as a cap
+(the exception is a message that is *one* enormous block, which is
+truncated, since dropping it would leave the row blank). A reply still
+streaming is **never** capped, because a row that stopped growing at two
+hundred lines while the model was plainly still writing reads as the
+stream having died. And the input's two blocks share **one** "Show all",
+since they are two halves of one answer -- while input and output have
+their own, since wanting the whole of a `new_string` says nothing about
+wanting the whole of the build log under it.
+
+**The Compose app does not wrap raw text any more**, per your call: a
+tool's leftover input fields and its output pan sideways like the command
+already did. A wrapped log destroys the column alignment that carried its
+meaning, one line at a time and only on the long lines -- so iris was
+right and Compose was the one to change.
+
+## 2026-09-08: redrawing one widget cost O(its own primitives squared)
 
 Your report -- expanding a tool card with a long horizontally-scrolling
 edit in it lags -- is a framework defect, not a text-layout one, and the
@@ -39,21 +88,15 @@ where that slot's handle sits in its owner's `ActiveData::primitives`
 
 `benches/message_list.rs` grew scenario **(g)** for it, and the number to
 read is per-glyph: flat as N grows is the pass condition, and a total
-hides it. Two things about that file: it also caught the quadratic in
-`--phone`-sized text, and it had stopped running at all -- scenarios (a)
-and (e) built a `LazySpan` with no mask around it, which the span now
-asserts against, so the whole benchmark panicked on its second line.
-Fixed in the same change.
+hides it. That file had also stopped running at all -- scenarios (a) and
+(e) built a `LazySpan` with no mask around it, which the span asserted
+against, so the whole benchmark panicked on its second line. Worked around
+here and fixed properly in the entry above, which deletes the assert.
 
-**What this does not fix, and what I would do next.** An open card still
-shapes, rasterises and submits *every* glyph of its input and output, not
-the screenful you can see -- iris does not cull within a widget, and a
-tool card is the one place that bites, because an `Edit`'s `old_string`
-and `new_string` go onto the card whole. The output half is already capped
-(`OUTPUT_LINES`/`OUTPUT_BYTES` in `tool.rs`, 80 lines or 4 KiB behind a
-"Show all"); the *input* half has no cap at all, and that is the
-asymmetry to close. Wrapping the block, which you asked for, changes the
-shape but not this cost: the same glyphs are laid out either way.
+**What this does not fix**, and what the entry below closes: an open card
+still shapes, rasterises and submits *every* glyph of its input and
+output, not the screenful you can see. iris does not cull within a widget,
+so the only bound available is a cap on what goes in.
 
 ## 2026-09-08: one `ScrollController`, a `Scrollable` trait, and `Pin`