From that RFC:

0.0.0.0/8 - Addresses in this block refer to source hosts on "this"
network.  Address 0.0.0.0/32 may be used as a source address for this
host on this network; other addresses within 0.0.0.0/8 may be used to
refer to specified hosts on this network ([RFC1122], Section
3.2.1.3).

(note that it only says "source address")

which was based on RFC 1122, which states:

We now summarize the important special cases for Class A, B,
and C IP addresses, using the following notation for an IP
address:

    { <Network-number>, <Host-number> }

or
    { <Network-number>, <Subnet-number>, <Host-number> }

...

(a)  { 0, 0 }

This host on this network.  MUST NOT be sent, except as
a source address as part of an initialization procedure
by which the host learns its own IP address.

See also Section 3.3.6 for a non-standard use of {0,0}.

(section 3.3.6 just talks about it being a legacy IP for broadcasts - I don't think that even works any more)

it makes connecting to localhost as easy as http://0:8080/ (for port 8080, but omit for port 80).

The thing is that it's not supposed to work, so it's essentially relying on undefined behaviour. Typing [::1]:8080 is nearly as easy.

skimming through these PRs, at least for WebKit, I don't see tests for shorthand IPs like 0 (and no Apple device to test with). What are the chances they missed those..?

I haven't seen the PRs, but IP comparison should really be using the binary form of the IPv4 address (a 32-bit number), not the human-friendly form.