I can't give a short answer, because it would be super controversial. But I can give you a good long answer.
Firstly, to bring everyone reading this up to speed: Ping is a very specific network protocol. You might think it's a UDP or maybe TCP based thing, but it's not: Ping is based on ICMP, a set of messages designed for communications between routers. Specifically, PING is built around the ECHO ICMP message (message type #8, and the reply is message type #0). Other ICMP messages are things like type #11, "Time Exceeded" (sent when a packet comes in with no time remaining), or type #4, "Source Quench" used when there's too much data coming in.
Fun historical diversion: in the original design of the internet, the idea is that each site (with possibly multiple real computers) would connect to the "internet message processor" (IMP); the imps were then connected together via phone lines. The ICMP messages were designed to go between the IMPs and not between the actual computers. Many IMPs also have a dedicated teletypewriter connected to the imp. So there's networking protocols designed for programs to use to connect to programs, and other network protocols designed just for the networking hardware.
So right away there's a problem: the PING message is unlike every other type of packet that the network codes uses. It's not TCP, it's not UDP, it's not a higher level protocol like ATOM or RSS or HTTP that we might want to support. Adding PING support would actually have required work.
The second problem is that PING is (no offense) somewhat overrated. Yes, it's often used to tell if a remote server is up. On the other hand, ICMP messages are routinely suppressed by firewalls, aren't routed by routers, and dropped entirely going in and out of a data center. Worse, a server can be effectively dead for use, and yet because of where PING is handled, it's still sending out replies.
This means that if you rely on PING to determine that a server is unreachable or dead, you'll have a bad time: there are plenty of cases where the server is perfectly fine and usable, but PING won't find it. And the flip side is true: just because PING can connect to the server, it doesn't mean that the server will actually talk to you.
The general network programming recommendation is: just try to connect to the server. If it fails, then you know it failed. If it works, you know it works. But trying to guess from PING if a server if OK isn't nearly as effective as you want it to be.
Source: I was the program manager (PM) on the team at Microsoft that decided to not support PING in UWP apps.