tamer: tests/xmli: Estimate tamec time in milliseconds

Going higher than that doesn't make sense because we're in shell and
invoking commands all around this, so even milliseconds isn't going to be
entirely accurate here.  However, what I am more interested in is observing
time relative to other runs; this isn't intended for profiling, but for
eyeballing unexpected behavior.

DEV-13708

tamer/tests/xmli/test-xmli

@@ -40,15 +40,25 @@ timed-tamec() {
 
   local -i ret=0
 
-  command time -f "%Es %F/%Rfault %I/%Oio %Mrss %c/%wctx \n%C" -o "$dir/time.log" \
+  # /usr/bin/time does not have high enough precision
+  # This won't be wholly accurate because time is spent in shell,
+  #   but it'll be close enough.
+  local -i start_ns=$(date +%s%N)
+
+  command time -f "%F/%Rfault %I/%Oio %Mrss %c/%wctx \n%C" -o "$dir/time.log" \
     "${TAMER_PATH_TAMEC?}" -o "$dir/$out" --emit xmlo "$dir/$in" \
     &> "$dir/tamec-$out.log" \
     || ret=$?
 
+  local -i end_ns=$(date +%s%N)
+  local -i elapsed_ms=$(( (end_ns - start_ns) / 1000000 ))
+
   # First line will be "Command exited with non-zero status N" on failure.
   # The last line will be '%C' above.
   # So, we want the second-to-last line.
-  echo -n "$(tail -n2 "$dir/time.log" | head -n1)"
+  #
+  # Newline omission is intentional.
+  printf "<%3dms %s" "$elapsed_ms" "$(tail -n2 "$dir/time.log" | head -n1)"
 
   return "$ret"
 }