Expand all leading tabs in markdown files

Given how problematic leading tabs in markdown files can be,
we have expanded all leading tabs in markdown files by:

```sh
find * -type f -name '*.md' -print |
  while read -r file; do
    rm -f "$file.tmp" ;
    expand < "$file" > "$file.tmp" ;
    mv -f -v "$file.tmp" "$file" ;
  done
```

Given the previous changes to `bin/pandoc-wrapper.sh` and other recent
commits and recent pull requests, the were were **NO** changes to
any HTML files.  This is a **good thing** as the expression goes.

No there are no leading tabs in any markdown files.
A check to catch any re-introduction of leading tabs
into any markdown file is on the way.

BTW: We do not have anything against leading tabs in files,
even though we have had to deal with their impact on forming
HTML from markdown files (as well as the calc source code).
It is just that this is one of those cases where tabs conflict
(sometimes caused by people wanting to change the classic
8-space tab model) with a desired outcome.

Perform `make www` to perform the above and test under macOS.

2019/adamovsky/README.md

@@ -291,7 +291,7 @@ be applied.
 [6]: https://en.wikipedia.org/wiki/Lambda_calculus           "Lambda Calculus on Wikipedia"
 [7]: https://en.wikipedia.org/wiki/Combinatory_logic         "Combinatory Logic on Wikipedia"
 [8]: https://inazz.jp/unlambda/                              "Online Unlambda interpreter"
-[9]: ../../2014/birken/index.html			     "Best use of port 1701"
+[9]: ../../2014/birken/index.html                            "Best use of port 1701"
 
 <!--
 

2019/dogon/README.md

@@ -53,7 +53,7 @@ We truly have not much more to add, except two things:
   model RAM, you'd want to reduce the value of the Z define in the Makefile
   to `0x10000000`, otherwise the entry will promptly crash. To do this you can do:
 
-	make clobber MEMORY=0x10000000 all
+        make clobber MEMORY=0x10000000 all
 
   Notice that that value must be a power of two!
 

2019/endoh/README.md

@@ -93,7 +93,7 @@ Okay, check the backtrace.
     #5  0x00005555555557ce in x69 () at prog.c:105
     #6  0x0000555555555828 in x6e () at prog.c:110
     #7  0x0000555555555747 in x65 () at prog.c:101
-	...
+        ...
 ```
 
 See the line numbers and lookup the ASCII table.

2019/mills/README.md

@@ -532,8 +532,8 @@ assignments.  For example, the simple one-layer RNN could be defined like
 
 ``` <!---make-->
     -DNW=' x  = I(n), hp = I(128),                  \
-	   h  = C(hp, T(A(L(128, x), L(128, hp)))), \
-	   y  = L(n, h)'
+           h  = C(hp, T(A(L(128, x), L(128, hp)))), \
+           y  = L(n, h)'
 ```
 
 The network declares `x` as an input vector (there must be a declaration for
@@ -578,14 +578,14 @@ GRU network:
 ``` <!---make-->
     -DHS=128,                                 \
     -DNW=' x  = I(n),                         \
-	   y  = L(n, MD(MD(x)))'              \
+           y  = L(n, MD(MD(x)))'              \
     -DBK=' hp = I(HS),                        \
-	   z  = S(A(L(HS, x), L(HS, hp))),    \
-	   r  = S(A(L(HS, x), L(HS, hp))),    \
-	   c  = T(A(L(HS, x), L(HS, hp))),    \
-	   zc = OG(1, -1, z),                 \
-	   h  = C(hp, A(M(zc, hp), M(z, c))), \
-	   y  = h'
+           z  = S(A(L(HS, x), L(HS, hp))),    \
+           r  = S(A(L(HS, x), L(HS, hp))),    \
+           c  = T(A(L(HS, x), L(HS, hp))),    \
+           zc = OG(1, -1, z),                 \
+           h  = C(hp, A(M(zc, hp), M(z, c))), \
+           y  = h'
 ```
 
 We are using a few new tricks here -- first, we are defining `HS` as the size