All bazel fixed output derivations should be specific to the bazel
version that was used to generate them. There is not guarantee that the
build will still succeed or reproduces (without the cached fixed output)
if the fetch phase wasn't rerun with a different bazel version.
In the past bazel had been bumped but not all those packages that have
fixed outputs from bazel builds. This lead to compiling and somewhat
working TF versions that couldn't be reproduced without the cached fixed
outputs.
###### Motivation for this change
Update to latest upstream version
###### Things done
* [ ] Tested using sandboxing ([nix.useSandbox](http://nixos.org/nixos/manual/options.html#opt-nix.useSandbox) on NixOS, or option `sandbox` in [`nix.conf`](http://nixos.org/nix/manual/#sec-conf-file) on non-NixOS linux)
* Built on platform(s)
* [x] NixOS
* [ ] macOS
* [ ] other Linux distributions
* [ ] Tested via one or more NixOS test(s) if existing and applicable for the change (look inside [nixos/tests](https://github.com/NixOS/nixpkgs/blob/master/nixos/tests))
* [ ] Tested compilation of all pkgs that depend on this change using `nix-shell -p nix-review --run "nix-review wip"`
* [x] Tested execution of all binary files (usually in `./result/bin/`)
* [ ] Determined the impact on package closure size (by running `nix path-info -S` before and after)
* [ ] Ensured that relevant documentation is up to date
* [x] Fits [CONTRIBUTING.md](https://github.com/NixOS/nixpkgs/blob/master/.github/CONTRIBUTING.md).
###### Notify maintainers
cc @worldofpeace @tfc @jtraue
I am not sure if we still need the old packages, nothing explicitly
depends on polyml56 or polyml57 according to a grep, not sure if
external packages might (hol and isabelle depend on polyml, the latest
version).
New libffi doesn't have FFI_SYSV for x86/64 unix, this pulls in the
commit for the upstream version which fixes it, and ports that patch to
the 5.7 version. The 5.6 version is unchanged.
For ZHF: #80379
Also updates the matchable egg (used by egg2nix) from 1.0 -> 1.1 and
removes trailing slashes from the path prefix variables passed to
wrapProgram (they're unnecessary and only result in doubled-up slashes
in the values).
I din't try to pinpoint the exact commit, but we started getting:
> The extension smart is not supported for docbook
Reading pandoc docs, I can't see what use to us "smart" could be
when writing the in-between docbook (to be converted to html).
https://pandoc.org/MANUAL.html#extension-smart
Prior to this fix, changes to certain settings would not be applied
automatically and users would have to know to manually restart the
affected service. A prime example of this is
`services.mailman.hyperkitty.baseUrl`, or various things that affect
`mailman3/settings.py`
