ames: account for variable-size num fragments (#561)

Currently we assume that the number of fragments in a `meow` has a fixed
size of 4 bytes (which is also how we deserialize meows in in
[lull.hoon](https://github.com/urbit/urbit/blob/develop/pkg/arvo/sys/lull.hoon#L1397);
see `+sift-meow`), but in some cases we will retrieve from the cache a
meow that has less than 4 bytes, which means that we will go over the
actual length of the meow when calculating the size of the response to
be sent.

(To reproduce the problem, [this pdf](https://www.lab6.com/0) will cause
the publisher to crash on the last fragment if anybody requests it using
remote scry)

This PR fixes this by checking how many bytes are actually used in a
word, and then using that information to correctly calculate the size of
the `meow`.

Thanks to @pkova and @joemfb for helping with this.

pkg/vere/io/ames.c

@@ -112,7 +112,7 @@
   typedef struct _u3_meow {
     c3_y    sig_y[64];                  //  host signature
     c3_w    num_w;                      //  number of fragments
-    c3_w    siz_s;                      //  datum size (actual)
+    c3_s    siz_s;                      //  datum size (actual)
     c3_y*   dat_y;                      //  datum (0 if null response)
   } u3_meow;
 
@@ -358,12 +358,25 @@ _fine_peep_size(u3_peep* pep_u)
     pep_u->len_s);
 }
 
+static inline c3_y
+_fine_bytes_word(c3_w num_w)
+{
+  return (c3_bits_word(num_w) + 7) >> 3;
+}
+
 static inline c3_s
 _fine_meow_size(u3_meow* mew_u)
 {
+  c3_y cur_y = 0;
+  if (mew_u->siz_s != 0) {
+    cur_y = sizeof(mew_u->num_w);
+  }
+  else {
+    cur_y = _fine_bytes_word(mew_u->num_w);
+  }
   return (
     sizeof(mew_u->sig_y) +
-    sizeof(mew_u->num_w) +
+    cur_y +
     mew_u->siz_s);
 }
 
@@ -603,7 +616,9 @@ _fine_sift_meow(u3_meow* mew_u, u3_noun mew)
     //  parse number of fragments
     //
     u3r_bytes(cur_w, num_w, (c3_y*)&mew_u->num_w, mew);
+    num_w = c3_min(num_w, (len_w - cur_w));
     cur_w += num_w;
+    u3_assert(len_w >= cur_w);
 
     //  parse data payload
     //
@@ -735,10 +750,22 @@ _fine_etch_meow(u3_meow* mew_u, c3_y* buf_y)
   memcpy(buf_y + cur_w, mew_u->sig_y, sig_w);
   cur_w += sig_w;
 
-  //  write number of fragments
-  //
-  _ames_etch_word(buf_y + cur_w, mew_u->num_w);
-  cur_w += sizeof(mew_u->num_w);
+  {
+    c3_y num_y[4];
+    c3_y len_y = _fine_bytes_word(mew_u->num_w);
+
+    //  write number of fragments
+    //
+    _ames_etch_word(num_y, mew_u->num_w);
+    memcpy(buf_y + cur_w, num_y, len_y);
+
+    if (mew_u->siz_s != 0) {
+      cur_w += sizeof(mew_u->num_w);
+    }
+    else {
+      cur_w += len_y;
+    }
+  }
 
   //  write response fragment data
   //