fix(xlings): the remedy for a SubOS with no subos_info is a command that works

[Sunrisepeak]

Three messages tell the user how to get an undescribed SubOS described, and all three name something that does not do it. The wording has now been wrong in two opposite directions:

1. xlings self update — which does not touch a subos manifest at all.
2. "an existing one is NOT backfilled; recreate the SubOS" — the correction to (1), and also wrong.

xlings self doctor --fix has always described the ACTIVE subos. What it got wrong was the content: it stamped the block with whatever the built-in default libc happened to be, rather than with what the SubOS actually runs. On a measured 39-subos home that produced two subos declaring glibc@2.44 over a sysroot serving 2.39.

openxlings/xlings#553 (xlings 2026.8.17.1) fixes the content: the runtime comes from the recorded binding, the workspace record, or the payload behind lib/libc.so.6 — and when none of those can answer, the runtime key is omitted rather than filled with a guess.

No ordering constraint

self doctor --fix supplies the block on today's xlings too, so this message is true before that release lands; it just writes a runtime that may be wrong. The parenthetical says which version records it truthfully so the user can tell the two apart.

The omitted-runtime state already has a home here

runtime_binding.cppm already handles info.present && info.runtime.empty():

SubOS '{}' has no runtime identity in subos_info.runtime; runtime rules cannot be evaluated for artifacts built here

That was written for a hypothetical. It is now the honest state a described-but-unprovable SubOS reports, so nothing here needs to change to accept the new format — which is worth stating explicitly, since the alternative reading is that it was untested and happens to work.

Why the wording matters more than usual

"from inside this SubOS" is load bearing: doctor inspects the subos the run is in and no other. When mcpp is the one complaining, the user is in it.

And telling someone to delete and recreate a SubOS to recover an environment is a large ask — it was never the smallest thing that works.

Sites changed: platform/xlings/subos_info.cppm, platform/runtime_binding.cppm, toolchain/lifecycle.cppm.