Guess what?!?!
I took apart the scope and found the JTAG pin headers on the main board were gone. Just 10 unpopulated solder pads
DAMN 
I just went through the process of unlocking everything on my MSO1104Z-S here, and I found the same thing when I opened it up. I ended up soldering in a header, but it does take a little bit of work since you have to fully disassemble the scope. And no point bothering to keep the warranty sticker after that

However the process worked fine. I used an Altera USB Blaster clone I got from eBay, wired it up according the JTAG pinouts specified here:
https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/ug/ug_usb_blstr.pdfFor reference, here are the USB blaster pin assignments for JTAG mode:
| USB Blaster Pin | Signal |
| 1 | TCK |
| 2 | GND |
| 3 | TDO |
| 4 | VCC |
| 5 | TMS |
| 6 | N/C |
| 7 | N/C |
| 8 | N/C |
| 9 | TDI |
| 10 | GND |
When I was hooking it up I just ignored any signals that aren't on that list.
I used OpenOCD 0.9.0 on Windows. My scope had firmware 4.03.SP2 installed, and I still had some time left on the feature trial licenses so I just halted the processor while the trial time remaining screen was showing and dumped the image.
I got bored after about an hour and wanted to see if I was getting anything, so I stopped the process with about 16MB dumped and used rigup 0.4.1 (the mso1000z version, no patches or anything applied). It found the keys and the generated license worked fine.
As a side note, I briefly tried building and running rigup using the bash shell support in Windows 10, but it faulted with a failed assertion that I didn't look into. Instead I built it in Visual Studio and it ran fine with just a couple of very minor tweaks to change a couple of POSIX specific calls.