Flake follows: resolve all follows to absolute

[balsoft]

It's not possible in general to know in computeLocks, relative to
which path the follows was intended to be. So, we always resolve
follows to their absolute states when we encounter them (which can
either be in parseFlakeInput or computeLocks' fake input population).

Fixes #6013
Fixes #5609
Fixes #5697 (again)

[blaggacao]

Fixes #5386, too?

[balsoft]

Likely

[blaggacao]

Thank you very much for this! Joy and Acknowledgment.

Pre:

❯ nix flake lock
warning: Git tree '/home/blaggacao/src/github.com/input-output-hk/bitte-cells' is dirty
error: input 'std/alejandra/fenix/nixpkgs' follows a non-existent input 'alejandra/nixpkgs'
(use '--show-trace' to show detailed location information)

---

Post:

❯ nix flake lock
warning: Git tree '/home/blaggacao/src/github.com/input-output-hk/bitte-cells' is dirty
warning: creating lock file '/home/blaggacao/src/github.com/input-output-hk/bitte-cells/flake.lock'

[sheyll]

Yeah, thanks so much!

[ztzg]

Hi @balsoft, all,

Could we label this PR with backport 2.6-maintenance? The patch applies and works on top of that branch.

(Alternatively, I could open another PR with the backport, if that's better.)

Thanks, -D