Spend the presses on the open question, and report what learning actually folded

Two changes to the measurement, both about not wasting a finger.

The wrong-finger control is off by default now. Across every run this project
has made the wrong finger has matched zero times out of forty-odd taps; the
question that is still open is the false-negative rate, and each control tap
spends a press that could have measured it. Pass a count to put the control
back, and the summary says plainly when it was not run rather than printing a
zero that reads like a result.

And a trend run now reports how many frames learning folded, out of the daemon's
own transcript, alongside the container size. Without that a run cannot tell
"learning fired and did not help" from "learning never fired" -- which is exactly
the confusion that let the first trend run measure a static template three times
and look like a result.

Default trial length goes to fifteen taps, since dropping the control freed the
presses and ten was too few to separate three-in-ten from five-in-ten.

packaging/fplearn.sh

@@ -40,6 +40,19 @@ UNIT=fingerprintd-test
 BIN=/tmp/fingerprintd
 COMMON="--daemon --verbose --edge-wake --sfs-root=/var/lib/fingerprintd/sfs --sfs-writable --rpmb-write"
 
+# How many frames learning actually folded, out of the daemon's own transcript.
+# Without this a trend run cannot tell "learning fired and did not help" from
+# "learning never fired" -- which is exactly the confusion that made the first
+# trend run measure a static template three times.
+folds() {
+    L=$(ls -t /var/log/fingerprintd/*.log 2>/dev/null | head -1)
+    [ -n "$L" ] || { echo "  (no transcript)"; return; }
+    m=$(grep -c -- "-> MATCH fid=" "$L" 2>/dev/null || echo 0)
+    f=$(grep -c "folded in (metric" "$L" 2>/dev/null || echo 0)
+    s=$(grep -c "learn: template saved" "$L" 2>/dev/null || echo 0)
+    echo "  folds: $f frame(s) over $m matched press(es), $s save(s)"
+}
+
 sizes() {
     # A template is stored twice, the container and its backup, so the sizes
     # come in pairs. The BODY is the container minus its 4096-byte header.
@@ -115,7 +128,7 @@ enrol)
     echo; echo "next: fplearn.sh base" ;;
 
 base)
-    N=${2:-10}; W=${3:-5}
+    N=${2:-15}; W=${3:-0}
     echo "=== BASELINE: learning OFF ==="
     restart "--learn=0" || exit 1
     echo "sizes before:"; sizes
@@ -124,7 +137,9 @@ base)
     echo; echo "next: fplearn.sh trend" ;;
 
 trend)
-    N=${2:-10}; W=${3:-5}
+    # No wrong-finger taps by default: zero false accepts is settled over forty
+    # of them, and the presses are better spent on the false-negative rate.
+    N=${2:-15}; W=${3:-0}
     echo "=== TREND: learning ON, three runs on one template ==="
     restart "--learn=1" || exit 1
     echo "sizes at the start:"; sizes
@@ -132,7 +147,7 @@ trend)
     while [ $r -le 3 ]; do
         echo; echo "----- learning run $r of 3 -----"
         /tmp/fptrial.sh "$N" "$W"
-        echo "sizes after run $r:"; sizes
+        echo "after run $r:"; sizes; folds
         r=$((r+1))
     done
     echo