Tweaked amor.py to use size field for amortized measurements

This makes more sense when using benchmarks with sparse sampling rates.
Otherwise the rate of sampling also scales the resulting measurements
incorrectly.

If the previous behavior is required (if you want to ignore buffer sizes
when amortizing read/writes for example), the -n/--size field can always
be omitted.

scripts/amor.py

@@ -84,11 +84,9 @@ def main(csv_paths, output, *,
         size=None,
         fields=None,
         defines=[]):
-    # default to amortizing and per-byte results if size is present
+    # default to amortizing
     if not amor and not per:
         amor = True
-        if size is not None:
-            per = True
 
     # separate out renames
     renames = list(it.chain.from_iterable(
@@ -163,7 +161,7 @@ def main(csv_paths, output, *,
             # find amortized results
             if amor:
                 amors.append(r
-                    | {f: sums[f] / (j+1) for f in fields}
+                    | {f: sums[f] / size_ for f in fields}
                     | ({} if meas is None
                         else {meas: r[meas]+'+amor'} if meas in r
                         else {meas: 'amor'}))