refactor(daemon): split touch interaction orchestration into semantic modules (#1748)

* refactor(daemon): split touch interaction orchestration

Closes part of #1691: interaction-touch.ts becomes a router; press, fill,
direct-iOS, shared runtime, Android readiness, and response projection each
own one module. Behavior is unchanged.

* test(daemon): split touch interaction coverage by module

Redistributes all 87 discovered cases across the new module topology and
re-keys the two touch-family fallow baseline entries to the paths that now
hold the same (net one fewer) findings.

* docs: point ADR 0014 at the merged Android readiness regression file

* refactor(daemon): give targeted-touch admission its own module

Keeps interaction-touch-press.ts inside the 300-line budget after the
complexity decomposition: admission (surface/capability/button policy, target
parsing, @ref staleness and mutation admission) answers its own question.

* refactor(daemon): drop the redundant targeted-touch label alias

* test(daemon): split touch suites along the new production seams

Adds the press-admission suite the production module was missing and splits the
four over-budget suites along new production seams (direct-iOS eligibility,
Android ref freshness, touch payload). Every suite installs the full device mock
set: three Android-session cases regressed to TOOL_MISSING on a runner without
adb when the mock set was trimmed per file.

Author: thymikee

docs/adr/0014-session-ref-frame-lifetime.md

@@ -6,7 +6,7 @@ Accepted; implemented and enforced on `main` across all platforms. Fresh live ev
 every production seam except Android blocking-dialog recovery, which shipped as a documented,
 accepted gap: no deterministic repro exists to raise the system dialog on demand, and the seam's
 transition and abort logic are pinned by fixture regressions
-(`android-system-dialog-ref-frame.test.ts`, `interaction-android-recovery-abort.test.ts`). The
+(`android-system-dialog-ref-frame.test.ts`, `interaction-touch-android-readiness.test.ts`). The
 staged migration plan lives in this file's git history; PR #1241 was an earlier compatible
 transitional fix, not this architecture migration.
 

fallow-baselines/health.json

@@ -148,9 +148,24 @@
         "count": 1
       }
     },
-    "src/daemon/handlers/__tests__/interaction.test.ts": {
+    "src/daemon/handlers/__tests__/interaction-touch.test.ts": {
       "crap_moderate": {
-        "count": 6
+        "count": 2
+      }
+    },
+    "src/daemon/handlers/__tests__/interaction-touch-press.test.ts": {
+      "crap_moderate": {
+        "count": 3
+      }
+    },
+    "src/daemon/handlers/__tests__/interaction-touch-press-admission.test.ts": {
+      "crap_moderate": {
+        "count": 2
+      }
+    },
+    "src/daemon/handlers/__tests__/interaction-touch-response.test.ts": {
+      "crap_moderate": {
+        "count": 1
       }
     },
     "src/daemon/handlers/__tests__/record-trace.test.ts": {
@@ -215,11 +230,6 @@
         "count": 1
       }
     },
-    "src/daemon/handlers/interaction-touch.ts": {
-      "crap_moderate": {
-        "count": 3
-      }
-    },
     "src/daemon/handlers/record-trace.ts": {
       "complexity_moderate": {
         "count": 1