fix(#1520): the quickstart page is a landing source, so a README that SHRINKS into a pointer can land

[localai-bot]

check-doc-checkpoint refuses a README.md claim change that touches no member
of LANDING_SOURCE_FILES. The set names .agents/mission.md, CMakeLists.txt,
three benchmarks/demo/*.json files and the two example mains, and it has no
entry for a document that owns the start of a reader's path.

The gate's own message names the two failures it exists to stop:

The README is the landing page. Routine checkpoints belong in the
purpose-specific docs. Co-edited public projections never justify README
churn.

Both are a README that GROWS. row/DOCS-QUICKSTART-1281 is the opposite. It
adds docs/QUICKSTART.md and cuts the README ## Quickstart block from three
command fences to a four-line pointer at that page, so the README loses material
and the material moves to the purpose-specific document. At head 1b6e458c8
that branch reports 88 gates ok, zero skips and one failure, doc-checkpoint range, while touching no member of the set, which it honestly should not,
because nothing in .agents/mission.md became untrue.

Add docs/QUICKSTART.md to the set. Every existing member is something the
README quotes: the mission it paraphrases, the build entry point its build line
invokes, the demo measurements its numbers come from, and the two example mains
its commands run. The quickstart page is that relation with the direction made
explicit. The README's claim about where a reader starts changed BECAUSE the
page now exists and now owns the starting procedure, and the page is a source
rather than a projection, because nothing else in the tree records what it says.

This admits one exact path and no class. docs/BUILD.md, docs/ROCM.md,
docs/RELEASES.md and every public projection still cannot license a README
claim change, and landing_page still permits a README edit rather than
demanding one. Three cases in tests/scripts/test_doc_checkpoint.py state that:
the new one was RED before this change with the exact landing source refusal,
and the two guards were green before and after. Three mutations red them one at
a time, and the checker at origin/main refuses d69a43c51 while the checker
here accepts it.

Deliberately its own change. The previous implementer on the quickstart row
refused to widen the set inside the change it unblocks, because AGENTS.md names
making a red gate green that way as the forbidden move. That refusal was
correct, so the widening is made here with its own spec, its own red-first
evidence and its own mutations.

FOLLOWING_AGENTS_PROTOCOL

Following-Agents-Protocol: true
AI-Assisted: true
Assisted-by: Claude:Opus-5 [Claude Code]