racket: move to pkgs/by-name, combine full and minimal definitions

[TomaSajt]

Related: #392174

Why?

pkgs/by-name gets rid of the metadata role of the file paths. Now noone can complain about whether it is an interpreter or a compiler.

To streamline this migration, I also combined minimal.nix and default.nix into the package.nix file.

IMO this is much more readable than using overrideAttrs: everything is in one place so you know what's going on.

Even though we're in pkgs/by-name, we have to still use all-packages.nix to supply the correct stdenv and the boolean.

---

Note: this PR causes a rebuild of racket-minimal because a key set to an empty string made by lib.optionalString is technically different from the key being unset.

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandboxing enabled in nix.conf? (See Nix manual)
  • sandbox = relaxed
  • sandbox = true
• Tested, as applicable:
  • NixOS test(s) (look inside nixos/tests)
  • and/or package tests
  • or, for functions and "core" functionality, tests in lib/tests or pkgs/test
  • made sure NixOS tests are linked to the relevant packages
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage
• Tested basic functionality of all binary files (usually in ./result/bin/)
• 25.05 Release Notes (or backporting 24.11 and 25.05 Release notes)
  • (Package updates) Added a release notes entry if the change is major or breaking
  • (Module updates) Added a release notes entry if the change is significant
  • (Module addition) Added a release notes entry if adding a new NixOS module
• Fits CONTRIBUTING.md.

---

Add a 👍 reaction to pull requests you find important.

[rc-zb]

Is this a revocation of #375992? Actually, I made that because then I intended to integrate the Racket package system into Nixpkgs and bootstrap Racket from BC(minimal) to CS(minimal) to CS(full) instead of utilizing the bytecode built in the released source archives (see #373716). The separation was made so that I could focus on the minimal distribution. This plan was stuck due to obscure build failures and I have had no time to try to settle them, even until now. But I don't think it should be considered to be failed already, and someday I may pick it up.

If the classification is really important according to the upstream, IMO you should better move the recipes to pkgs/development/compilers/racket, instead of fitting them into the flattened by-name structure, which is more suitable for one-version-only packages. After all,

Even though we're in pkgs/by-name, we have to still use all-packages.nix to supply the correct stdenv and the boolean.

[drupol]

Suggested change

	++ lib.optional disableDocs "--disable-docs"
	++ lib.optionals disableDocs [ "--disable-docs" ]