Tidy up CRYPTO_EX_DATA structures.

opening the output file with "wb" to truncate it except on VMS
(where the file now keeps its original length because it is opened
with "rb+" -- does VMS have ftruncate?)

Slightly clarify the RAND_... documentation.

(we're now using fopen(..., "rb+") instead of fopen(..., "wb"),
so the file is not truncated automatically).

benefit of MS-DOS users.

non-function pointers to function pointers and vice versa.
The current solution is to have unions that describe the
conversion we want to do, and gives us the ability to extract
the type of data we want.

The current solution is a quick fix, and can probably be made
in a more general or elegant way.

((void *)0), by have a 0 instead.

which does no harm fixes that problem.

that are directly in crypto/, and prepare for a possible disabling of
certain messages that DEC C spews out.

are not guaranteed to be unpredictable.

(and rename a target in test/Makefile.ssl to make it
easier to guess the name of the file executed by it)

"key" is NULL.

  because otherwise BN_rand will fail unless DEVRANDOM works,
  which causes the programs to dump core because they
  don't check the return value of BN_rand (and if they
  did, we still couldn't test anything).

- add comment to some files that appear not to be used at all.

and re-enable seeding with more data than read from DEVRANDOM -- just
don't pretend it contains entropy.

(only the CRYPTO_push_info's in the apps/ directory
are included in the CVS commit, not all those I used
in crypto/)

returns int (1 = ok, 0 = not seeded). New function RAND_add() is the
same as RAND_seed() but takes an estimate of the entropy as an additional
argument.