If the curl fails or receives a non-2xx status code, it retries, backing
off a little bit, until it either succeeds, or exceeds five attempts.

Adds support for --prop-file, --prop-url and --prop-name, which let one
augment the boot sequence of a ship with additional events.

Back-ports the vere changes from urbit/urbit#5470 onto current vere. See
that for additional details/history. This is a mostly straightforward,
copy-paste port of the changes from that original PR, with the following
exceptions:
- Boot sequence construction now happens in _pier_boot_make.
- We don't modify the way the default boot sequence is acquired (yet).

We will want to follow up on the latter, eventually.

The %autoprop "prop builder" that was included in #5470 should still
operate fine and be useful for automating keeping certain kinds of props
up to date. (Save for minimal kelvin compatability disrepair it has
fallen into due to the flow of time.)

Previously running `(swp 3)` on a 70MB atom would always exhaust a 2gb
loom. With this we can easily `(swp 3)` and even `(swp 0)` a 500MB atom.
Note that fares a lot better even with no special casing of blocksizes
0, 3 and 5.

Not sure what you want to do with this, but on Sonoma, clang 15.0.0 is
what is installed.

I'm not currently able to build on my OSX machine, seemingly because I
can't get bazel and my nix libtool package to cooperate.

For safety.

1.1.1t is 404ing on openssl.org.

Reverts #579

We have to revert this until @jalehman gets back and we can access
macstadium. CI on develop is broken otherwise. No biggie, we'll just put
this back after.

1.1.1t is 404ing on openssl.org.

Resolves #

See urbit/urbit#6802 and
urbit/urcrypt#8

Depends on #524

I can't think of a reason `MODULE.bazel` and `MODULE.bazel.lock`, both
of which are auto-generated, shouldn't be in the `.gitignore` file. I
keep manually ignoring them.

Fixes #640

Resolves #679.

#622 helped with this, but it's still overly laconic about what's going
on. I think this is better.

Informal pinging has been broken since #623, luckily this was
unreleased, but existed edge.

The PR replaces vere's current (buggy, insufficient) lockfile mechanism
with an advisory `fcntl()` write lock (`F_SETLK`, `F_WRLCK`) on the
first byte of vere's PID file (`$pier/.vere.lock`). It does not (yet?)
include the functionality to kill the lock-holder and take over the
pier, as I'm not sure that's desirable.

There have been many problems with vere's current lockfile
implementation. The biggest advantage of this approach is that the lock
is automatically released if the process is killed or the host system
crashes. The biggest question is: does it work reliably across docker
shared volumes, virtual machines, &c.