Add LastPostCode component details to SP5 host CPU

[mkeeter]

Fixes #2244

We would like to be able to read the LAST_POST_CODE register from the FPGA. This PR adds it as a ComponentDetail to the SP5_HOST_CPU component.

The MGS side is oxidecomputer/management-gateway-service#451, so this is marked as a draft (but can still be reviewed).

There are two interesting changes:

• Inventory::num_component_details now returns 1 for the SP5_HOST_CPU component
• Inventory::component_details now accepts a closure to handle any components in OUR_DEVICES

The rest of the PR is mostly plumbing, e.g. adding modules to the generated FMC code (because otherwise we have name collisions between espi and sequencer registers).

I tested this on the bench with a Grapefruit, and it "works":

➜  management-gateway-service jj:(zq) cargo run -pfaux-mgs -- --interface=en10 component-details sp5-host-cpu
   Compiling gateway-messages v0.1.0 (/Users/mjk/oxide/management-gateway-service/gateway-messages)
   Compiling gateway-sp-comms v0.1.1 (/Users/mjk/oxide/management-gateway-service/gateway-sp-comms)
   Compiling faux-mgs v0.1.1 (/Users/mjk/oxide/management-gateway-service/faux-mgs)
    Finished `dev` profile [unoptimized + debuginfo] target(s) in 7.01s
     Running `target/debug/faux-mgs --interface=en10 component-details sp5-host-cpu`
Sep 25 15:41:12.950 INFO creating SP handle on interface en10, component: faux-mgs
Sep 25 15:41:12.954 INFO initial discovery complete, addr: [fe80::c1d:8cff:fec0:e208%26]:11111, interface: en10, socket: control-plane-agent, component: faux-mgs
LastPostCode(LastPostCode(0))

I've also tested the Sequencer.last_post_code function on a Cosmo through humility hiffy. I would like to do an end-to-end test, but Cosmos with network connectivity are in high demand (see https://github.com/oxidecomputer/meta/issues/761)

[mkeeter]

I tested this on London without any issues:

Sep 26 17:33:27.145 INFO creating SP handle on interface london_sw0tp0, component: faux-mgs
Sep 26 17:33:27.146 INFO initial discovery complete, addr: [fe80::aa40:25ff:fe04:402%3]:11111, interface: london_sw0tp0, socket: control-plane-agent, component: faux-mgs
LastPostCode(LastPostCode(3925950469))
matt@castle ~ $ ./faux-mgs --discovery-addr='[fe80::aa40:25ff:fe04:402%3]:11111' --interface=london_sw0tp0 component-details sp5-host-cpu
Sep 26 17:34:07.478 INFO creating SP handle on interface london_sw0tp0, component: faux-mgs
Sep 26 17:34:07.480 INFO initial discovery complete, addr: [fe80::aa40:25ff:fe04:402%3]:11111, interface: london_sw0tp0, socket: control-plane-agent, component: faux-mgs
LastPostCode(LastPostCode(3925938431))

These are 0xea014005 and 0xea0110ff, which are both reasonable values.

The image is in /staff/dock/london/cosmo-sp-test and is flashed on BRM13250012.

[labbott]

LGTM I'll give it another look when you are ready for final review

[mkeeter]

@labbott Final review is pending oxidecomputer/management-gateway-service#451 ; I tagged @jgallagher but you're welcome to look as well!