updated the bar charts with the new benchmark data

doc/html_generated/benchmarks.html

@@ -196,6 +196,8 @@
 
 ----------
 
+The bar charts were generated using [**this google spreadsheet**](https://docs.google.com/spreadsheets/d/1p3MAURUfPzKT7gtJOVuJU2_yVKSqkoD1nbypA1K3618) by pasting the data from the tables.
+
 If you want a benchmark that is not synthetic - check out [**this blog post**](http://baptiste-wicht.com/posts/2016/09/blazing-fast-unit-test-compilation-with-doctest-11.html) of [**Baptiste Wicht**](https://github.com/wichtounet) who tested the compile times of the asserts in the 1.1 release with his [**Expression Templates Library**](https://github.com/wichtounet/etl)!
 
 While reading the post - keep in mind that if a part of a process takes 50% of the time and is made 10000 times faster - the overall process would still be only roughly 50% faster.

doc/markdown/benchmarks.md

@@ -191,6 +191,8 @@ In these particular cases **doctest** makes 0 allocations when the assert doesn'
 
 ----------
 
+The bar charts were generated using [**this google spreadsheet**](https://docs.google.com/spreadsheets/d/1p3MAURUfPzKT7gtJOVuJU2_yVKSqkoD1nbypA1K3618) by pasting the data from the tables.
+
 If you want a benchmark that is not synthetic - check out [**this blog post**](http://baptiste-wicht.com/posts/2016/09/blazing-fast-unit-test-compilation-with-doctest-11.html) of [**Baptiste Wicht**](https://github.com/wichtounet) who tested the compile times of the asserts in the 1.1 release with his [**Expression Templates Library**](https://github.com/wichtounet/etl)!
 
 While reading the post - keep in mind that if a part of a process takes 50% of the time and is made 10000 times faster - the overall process would still be only roughly 50% faster.