fix more broken markdown links

We fix more markdown links in markdown files that refer to files
internally to the web site.

Added tmp tools to find and print various markdown links:

* `tmp/md-links.sh` - print all markdown links
* `tmp/md-token-links.sh` - print all markdown links that start with a **%%TOKEN%%**
* `tmp/md-local-links.sh` - print all markdown links to a local file
* `tmp/md-link-paths.sh` - print local file paths referenced in markdown links

Fixed a number of markdown formatting problems with `thanks-for-help.md`.
Fixed markdown links to files in `thanks-for-help.md`.  Improved
heading levels in `thanks-for-help.md`, and made entry sections consistent.

Changed all markdown links that end in "))" to end in a better way.

Explained in `markdown.md` why we are avoiding markdown links that
end in "))".

Ran `make www` to perform the above.

TODO: Fix stuff we broke by mistake doing the above.

TODO: Fix more broken markdown links.

2019/mills/README.md

@@ -246,7 +246,7 @@ training loss will give you an idea of how well the network is learning and
 can let you know if the network is [overfitting or underfitting][7].
 
 As part of the training process, the data set (which for OMLET is the
-[Shakespeare.txt](Shakespeare.txt) file you gave on the command line) is divided into
+[Shakespeare.txt](Shakespeare.txt)file you gave on the command line) is divided into
 [training and validation sets][8] (by default, 95% of the data is used for
 training, but like most OMLET parameters, you can change this at compile-time).
 

2019/mills/index.html

@@ -578,7 +578,7 @@ validation loss is about 1.72. Comparing the validation loss and the
 training loss will give you an idea of how well the network is learning and
 can let you know if the network is <a href="https://en.wikipedia.org/wiki/Overfitting">overfitting or underfitting</a>.</p>
 <p>As part of the training process, the data set (which for OMLET is the
-<a href="Shakespeare.txt">Shakespeare.txt</a> file you gave on the command line) is divided into
+<a href="Shakespeare.txt">Shakespeare.txt</a>file you gave on the command line) is divided into
 <a href="https://en.wikipedia.org/wiki/Training,_validation,_and_test_sets">training and validation sets</a> (by default, 95% of the data is used for
 training, but like most OMLET parameters, you can change this at compile-time).</p>
 <p>At the end of the validation run, OMLET writes out a checkpoint file with a