Fix error in CP437 0x96 (small U with circumflex) translation to UNICODE
(typo: 0x00FB not 0x00F8).

Fixed one typo (LATIN_LATIN).

Extract the term size method used by SSH and pty into a function, and remove
the need to change the video mode.

always uses font zero now.

things.

This allows full control over the display, and avoids font mangling.

a struct vmem_cell.

This allows modifying a savescreen() rather than needing to vmem_gettext()
stuff, but also change things without hacking up the pixels array yourself.

Also, the vmem is an array of struct vmem_cell.

problems loading font data in SyncTERM over Telnet.

<Deuce> Easiest fix would be to have outbuf_append adjust outlen to be "long enough" if doubling isn't.
<Deuce>         /* Not enough room, double the size. */
<Deuce>         *outlen *= 2;
<Deuce>         if(*outp - *outbuf + len >= *outlen)
<Deuce>                 *outlen = *outp - *outbuf + len + 1
<Deuce> So yeah, blindly doubling won't work for utf-8.
<Deuce> Because a line will have more bytes than the number of characters in a line.

cvstat.

Per RFC-5198:
   2.  In Net-ASCII, CR MUST NOT appear except when immediately followed
       by either NUL or LF, with the latter (CR LF) designating the "new
       line" function.  Today and as specified above, CR should
       generally appear only when followed by LF.  Because page layout
       is better done in other ways, because NUL has a special
       interpretation in some programming languages, and to avoid other
       types of confusion, CR NUL should preferably be avoided as
       specified above.

It's not worth breaking all connections for some broken telnet servers.
Especially since we do not explicitly want CR behaviour which is the only
place it's allowed.

Per RFC-854:
   the sequence "CR NUL" must be used where a carriage return alone is
   actually desired;

It's quite possible that these "telnet servers" are actually raw socket
servers and should use raw sockets rather than telnet.

Note that conn_pty does not support PETSCII or ATASCII as the terminal
type since it sets TERMCAP, and I don't have a termcap entry for either of
those.

rather than no char, when calculating word lengths. We needed that 'continue'
here.

on cterm.

codes.

I like the -W @-code format modifier better for "WIDE" than for "WRAP" (no
body is using -W yet, right?) - so, change wrap to "->" and double-wide to
"-W".

terminal "type" string (ANSI, RIP, PETSCII, or DUMB).
Expose this value in JS via the new console.type read-only property.
The TERM @-code now expands to this string rather than the (often
client-supplied) sbbs_t.terminal string value.

dereference. (found by Coverity)

Expression syntax in function unicode_to_cp437
:-(

I took a short-cut with some of the names - TODO: fix. <sigh>

(e.g. "@code:BIG-RED@", the argument would get chopped to just "BIG")
so don't support format modifiers when a ':' is present in the @-code.

When appended to an @-code, displays it double-wide using "fullwidth" Unicode
code points if possible, spaces between the characters otherwise.

The pipe char is problematic to discuss in msg bases and on the Wiki. <shrug>

Create a sbbs_t method: wide(), which displays a string using "fullwidth"
Unicode characters, when possible, otherwise, double-spaced.
Exposed in JS via console.wide() and a new @-code: WIDE:<string>