docs: use a real chip name in the uvx example, not a <chip> placeholder

Qodo caught this on #128. `<chip>` inside a bash block is not an inert
placeholder — the shell splits it into two redirections:

  $ cd /tmp/cn && : > chip
  $ bash -c 'echo ARGS: defib burn -c <chip> -p /dev/ttyUSB0 -t'
  $ cat ./-p
  ARGS: defib burn -c /dev/ttyUSB0 -t

`<chip` redirects stdin, then `>` with the following word redirects
stdout into a file literally named `-p`. What defib actually receives is
`-c /dev/ttyUSB0 -t`, so the chip name becomes the port. Without a file
named `chip` in the cwd it fails "chip: No such file or directory"; with
one it runs the wrong command and drops a junk `-p` file that is
awkward to delete.

Use hi3516ev300, matching every other example in the README, and say to
substitute. The web/index.html line is unaffected — it interpolates the
selected chip before display.

Author: openipc-ai

README.md

@@ -21,10 +21,10 @@ uv tool install git+https://github.com/OpenIPC/defib
 pipx install git+https://github.com/OpenIPC/defib
 ```
 
-To run it once without installing anything:
+To run it once without installing anything (swap in your own chip and port):
 
 ```bash
-uvx --from git+https://github.com/OpenIPC/defib defib burn -c <chip> -p /dev/ttyUSB0 -t
+uvx --from git+https://github.com/OpenIPC/defib defib burn -c hi3516ev300 -p /dev/ttyUSB0 -t
 ```
 
 ## Quick Start