More attempts of reducing `flake.lock` file size

[jaredponn]

Following the experiments of #190, I felt inspired to go further. In #190, I noticed that CTL was the major culprit of too many nixpkgs, and in particular, their use of an older db-sync seemed to be the largest contributing factor. As such, this was patched in flake-lang here

Before this PR:

$ time nix flake lock

real    0m10.303s
user    0m10.008s
sys 0m0.178s

[07:22 PM] [jared@pletbjerg ~/Documents/Work/lambda-buffers]$ cat flake.lock | jq '.nodes | keys' | awk -e 'BEGIN { i = 0 } /nixpkgs/ { i++ } END {print i}'
351

This PR:

$ time nix flake lock

real    0m1.152s
user    0m1.112s
sys 0m0.035s

$ cat flake.lock | jq '.nodes | keys' | awk -e 'BEGIN { i = 0 } /nixpkgs/ { i++ } END {print i} '
149

Before merging

Before merging this, let's get https://github.com/mlabs-haskell/flake-lang.nix/pull/31 merged, then we can just refer to

    flake-lang.url = "github:mlabs-haskell/flake-lang.nix";

instead of

    flake-lang.url = "github:mlabs-haskell/flake-lang.nix?ref=jared/add-follows-to-nix-flake";