[staging] boost: default to boost 1.77 and refactor

[hjones2199]

Motivation for this change

Most packages that depend on the boost libraries in nixpkgs are pointing at the unversioned "boost" package. Unversioned boost is currently 1.69.0, which was released in 2018. The default should be the newest boost library available (1.77). I expect build-time breakages, but it should be fairly straightforward to pin broken packages to an older boost revision once they are identified.

The level of divergence (particularly the addition of boost-build) since I submitted #120259, and the fact that I wanted to combine the addition of boost 1.77 into the same PR made it easier to start the request from scratch.

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandbox = true set in nix.conf? (See Nix manual)
• Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review wip"
• Tested execution of all binary files (usually in ./result/bin/)
• 21.11 Release Notes (or backporting 21.05 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.

[hjones2199]

Do to the addition of patches in boost-build, boost 1.77 currently fails in the patch phase. Solving that without breaking old versions of boost on darwin is needed.

[hjones2199]

Replacing the darwin-specific patch with an equivelent substituteInPlace allows it to work across versions (including 1.77). Still probably a ton of broken packages that wont build without passing boost = boost16x but the PR is ready for review.

[r-burns]

uhd fails but is fixed by bumping to latest 4.1.0.3. I've tried a handful of packages and that's the failure I found so far

[oxalica]

What's blocking this issue now? Is it just uhd?

[trofi]

arc_unpacker now fails it's own tests in staging-next. Bisected down to eca5088 boost: default to boost 1.77:

$ nix build -f. arc_unpacker -L
...
arc_unpacker-unstable> /build/source/tests/dec/cat_system/int_archive_decoder_test.cc:38
arc_unpacker-unstable> ...............................................................................
arc_unpacker-unstable> /build/source/tests/test_support/decoder_support.cc:35: FAILED:
arc_unpacker-unstable>   {Unknown expression after the reported line}
arc_unpacker-unstable> due to unexpected exception with message:
arc_unpacker-unstable>   Premature end of file

[veprbl]

@trofi Right, that's #151469