fix(e2e): harden activation-e2e newer flows (issue #104) (#105)

Root-caused directory-picker's directory-row-frontend failure: all four
in-app @gorhom/bottom-sheet sheets (DirectoryBrowserSheet, DirectorySwitcher,
ModelPicker, VariantPicker) provide static percentage snapPoints but rely on
v5's enableDynamicSizing default (true), which never resolves without content
wrapped in a size-reporting component — so useAnimatedDetents() permanently
early-exits and the sheets can never actually open. Set
enableDynamicSizing={false} on all four (they already have explicit
snapPoints, so dynamic sizing was never needed).

variant-picker's chip failure was a stale test assumption: src/lib/
model-selection.ts's chooseModelSelection() deliberately returns null for a
fresh session (issue #37/#35 — the provider registry default is unreliable),
so a brand-new session has no model selected and the reasoning-effort chip
has nothing to key off of. Added testIDs (model-chip, model-option-*) and
updated the flow to explicitly pick a model first, matching real usage.

diff-scroll's missing markdown text: switched src/components/markdown/
Markdown.tsx from react-native-marked's FlatList-based default export to its
useMarkdown() hook rendered into a plain View. The chat screen already nests
this inside its own *inverted* FlatList (one row per message) — a nested
VirtualizedList inside an inverted outer list is a known RN footgun where the
inner content can render at zero height instead of just warning. We already
forced scrollEnabled:false + a large initialNumToRender, defeating
virtualization anyway, so rendering the parsed blocks directly loses nothing.
Extended the existing react-native-marked .d.ts shim (added for a React
18/19 ReactNode mismatch) to also declare useMarkdown/useMarkdownHookOptions.

Added diagnostic screenshots to directory-picker.yaml and diff-scroll.yaml
at the previously-failing steps for faster triage if these regress again.

Closes #104.

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

.maestro/flows/diff-scroll.yaml

@@ -80,8 +80,17 @@ name: "DiffView + CodeBlock horizontal scroll - wide diff/code line reachable by
 # render both the tool card and the markdown code block.
 - assertVisible:
     text: "Edit wide_diff_target.ts"
-- assertVisible:
-    text: "Here is a wide code sample"
+# Diagnostic screenshot (issue #104): the tool-card title above renders via
+# plain Text, while the paragraph below renders via react-native-marked's
+# nested FlatList inside this screen's own *inverted* FlatList. If this ever
+# regresses again, check whether this frame shows blank space between the
+# message header and the tool card (nested-FlatList-in-inverted-list content
+# not laying out) vs. genuinely missing/incorrect text.
+- takeScreenshot: diffscroll-S1b_before_markdown_text_check
+- extendedWaitUntil:
+    visible:
+      text: "Here is a wide code sample"
+    timeout: 20000
 - takeScreenshot: diffscroll-S2_history_loaded
 
 # --- CodeBlock: the wide fenced code block is already expanded (no tap

.maestro/flows/directory-picker.yaml

@@ -62,6 +62,10 @@ name: Directory picker - browse server folders and create a session in one
     id: "browse-folders-button"
 - assertVisible:
     text: "Browse Folders"
+# Diagnostic screenshot before the (previously flaky, see issue #104) wait
+# below — if this ever regresses again, compare this frame against
+# dirpicker-S2 to see whether the sheet actually opened.
+- takeScreenshot: dirpicker-S2b_browser_sheet_opened
 - extendedWaitUntil:
     visible:
       id: "directory-row-frontend"

.maestro/flows/variant-picker.yaml

@@ -13,6 +13,14 @@ name: Variant picker - reasoning-effort chip renders, selects, and still sends
 # on the runner host BEFORE this flow runs (same fresh, non-seeded normal-mode
 # server used by directory-picker.yaml — GET /provider's mock-model carries
 # low/medium/high variants; see tests/fixtures/mock-opencode-server.ts).
+#
+# A brand-new session has NO explicit model selection: src/lib/model-selection.ts
+# chooseModelSelection() deliberately returns null until the user picks one
+# (issue #37/#35 — the provider registry's "default" model is unreliable), so
+# app/session/[id].tsx's currentModelVariants is undefined and the chip does
+# NOT render on session open. This flow explicitly opens the model picker and
+# selects the mock provider's model first, which is what actually makes
+# variants available — matching how a real user reaches this chip.
 
 - launchApp:
     clearState: true
@@ -47,6 +55,18 @@ name: Variant picker - reasoning-effort chip renders, selects, and still sends
       id: "chat-message-input"
     timeout: 15000
 
+# Explicitly pick the mock provider's model — a fresh session has no model
+# selection yet (see comment above), so the variant chip has nothing to key
+# off of until this happens.
+- tapOn:
+    id: "model-chip"
+- extendedWaitUntil:
+    visible:
+      id: "model-option-mock-mock-model"
+    timeout: 10000
+- tapOn:
+    id: "model-option-mock-mock-model"
+
 # The chip only appears once catalog.load() (triggered on connect, see
 # app/_layout.tsx) has resolved GET /provider and found variants for the
 # selected model — wait rather than assert immediately.