I wanted to make CVM's forth use the Grid subsystem, but doing so
would break the stage binary. Hence, this split.

It isn't used in there anymore. Moved it to VE, the only place
where it's used.

If a line has a char below 0x20, we consider the line ended, we
stop the EMIT loop for this line and spit our NL.

This makes LIST stop wasting rows in environments with a tight
screen.

There are many situations where it can be useful. Worth it.

This allows us to remove AMOVE* words.

This would be useful, for example, to allow the assembler to write
straight to an AT28 EEPROM without going to RAM. This would be a
life saver in machines with tight RAM such as the SMS.

With KEY and EMIT being switch words, most of the high layer can
be defined before drivers.

In addition to this change, I've compacted core blocks which were
becoming quite sparse.

This allows us to remove one layer of indirection in (wnf). This is
possible now that EMIT is a switch word.

Slightly larger binary, but clearer code and slightly faster.

I'm pretty happy about how lightweight the implementation turns
out to be.

I think that when I added NL, I had troubles having access to CRLF's
address at boot time, which is why I had this system. But now that
CRLF is easily accessible during BOOT, we can simplify.

(and that will help us in the hopefully-upcoming change, which is
quite nice...)

They were only used in the ti84 recipe and were not worth their cost.

Adding a delay such as the 20ms one we have in AVR programmer's
initialization routine is tricky without a word like TICKS.

This implementation is highly inaccurate, but more accurate and
reliable than a "ballpark" DO..LOOP...

The few extra bytes they save in the core aren't worth the extra
complexity. This was initially done in a context where I had
troubles keeping the RC2014 binary with SDC inside the 8K limit.

At this point, even with the few extra bytes we add here, we're at
7200 bytes, so I'd say we're fine.

This marks the end of wordrefs in stable ABI. Nice.

The ":" now takes care of scanning for ";". Conceptually, having
";" as an immediate word is slightly simpler than the approach in
this commit, but when bootstrapping is involved, this simpler
approach gets murkier.

Moreover, it got even murkier-er when trying to de-stabilize EXIT,
so here we are.

This duplicated feature existed because of bootstrapping issues
with LIT", but again, with careful threading, we can clean things
up.

We can now have a proper "Collapse OS" prompt :)

I've mistakenly broken it in previous commit.

When writing the xcomp documentation, I realized that with careful
threading and by accepting a bit of code duplication in the xcomp
toolset, I could de-stabilize a couple of words.

(n) and EXIT are a bit trickier, but I think it can be done. It
would be nice to get rid of stable wordrefs...

The 1 byte limitation has been effective for a while now, but I
hadn't made the move yet, I wanted to see if the limitation would
cause me problems. It doesn't.

Doing this now slightly facilitates the IY->BC move in z80.

Bootstrapping: if you try to recreate the CVM binary from the
previous commit with this code, you'll have bootstrapping problems.
The first bootstrap will compile a binary with 2-bytes wide cells
but branching conditionals that yields 1-byte cells. That's bad.

I got around the issue by temporarily inserting a "397 399 LOADR"
instruction in cvm/xcomp.fs, right before the xcomp overrides. This
way, I force 1-byte cells everywhere on the first compiliation,
which then allows me to apply the logic change in cvm/vm.c and have
a properly running binary.

There is now no more actual code in stable ABI, only references.
This makes refactoring of this code much easier. For example,
changing IY to BC as the IP register.

Only its jump addr remains.

Only its jump at 0x33 remains.

I've also fixed a strange offset oddity in 8086's (n) placement.
It was off by 2, but strangely, it ran properly. Anyway, now it's
fixed.

Previously, it was impossible to cross-compile Collapse OS from a
binary-offsetted Collapse OS because stable ABI wordrefs would have
a wrongly offsetted address.

This solves the problem by replacing those wordrefs by direct,
hardcoded stable ABI offset references.