It wasn't used much, so I replaced its use in the kernel with direct code
and moved the routine in apps/ed, the only other place where it was used.

That's my mega-commit you've all been waiting for.

The code for the shell share more routines with userspace apps than with kernel
units, because, well, its behavior is that of a userspace app, not a device
driver.

This created a weird situation with libraries and jump tables. Some routine
belonging to the `kernel/` directory felt weird there.

And then comes `apps/basic`, which will likely share even more code with the
shell. I was seeing myself creating huge jump tables to reuse code from the
shell. It didn't feel right.

Moreover, we'll probably want basic-like apps to optionnally replace the shell.

So here I am with this huge change in the project structure. I didn't test all
recipes on hardware yet, I will do later. I might have broken some...

But now, the structure feels better and the line between what belongs to
`kernel` and what belongs to `apps` feels clearer.

The goal is to avoid mixing those routines with "character devices"
(acia, vpd, kbd) which aren't block devices and have routines that
have different expectations.

This is a first step to fixing #64.

Found via `codespell -q 3 -S ./tools -L splitted`

That was an interesting bug. It didn't cause a problem in emulation, but
in an RC2014 on an SD card, an include that didn't end with two newlines
would cause an infinite loop.

By not waiting until the SD card wan't busy any more after a write, we
exposed ourselves to errors if another operation was attempted right
after it.

With fsGetC becoming a "random address" API, it broke pgm. This commit
fixes it. To avoid adding the weight of a blkdev in pgm, I manage the
read offset directly in pgm.

Previous refacoring broke all seek/tell within fs. fs handles now lost
the responsibility to keep track of current position. It's blkdev's job.

This huge refactoring remove the Seek and Tell routine from blockdev
implementation requirements and change GetC and PutC's API so that they
take an address to read and write (through HL/DE) at each call.

The "PTR" approach in blockdev implementation was very redundant from
device to device and it made more sense to generalize. It's possible
that future device aren't "random access", but we'll be able to add more
device types later.

Another important change in this commit is that the "blockdev handle" is
now opaque. Previously, consumers of the API would happily call routines
directly from one of the 4 offsets. We can't do that any more. This
makes the API more solid for future improvements.

This change forced me to change a lot of things in fs, but overall,
things are now simpler. No more `FS_PTR`: the "device handle" now holds
the active pointer.

Lots, lots of changes, but it also feels a lot cleaner and solid.

I'm about to change that structure and I don't want fs to be messed up
by this. I'm proceeding step by step...

The ability to specify "0" routines in blkdev table is not used anymore
now that stdio is a separate subsystem.

Also, I'm preparing a blockdev refactoring and this complicates my work.

Using HL/DE was awkward and error-prone.

This allows us to break through the 64K limit for includes CFS in zasm,
a limit we were dangerously close to breaking. In fact, this commit
makes us go over that limit. Right in time!

Let go of that "meta os" thing. it's not as meta as I made it sound
like. It's a kernel.

When B>1, fsNext failed because HL was changed in the Seek call.

Also, clarify the role of recipes.

Very coarse, but I got them to work.

This way, we can interact with the mounted FS even when we change the
selected blkdev.

This allows more than one file to be allocated. `fls` properly lists
multiple files.

Very incomplete, but if you play your cards right, you can, with the
shell:

1. Create a new CFS
2. Mount it
3. Allocate a file with an arbitrary name
4. Have it listed with `fls`