One site for the loaders graph-ownership preamble

Every post-capture writer into the fact store opens with the same graph-ownership preamble: read store_graph.base_dir for the graph, mint the minimal anchor row where no capture ever reached this store under the name, and refuse to touch a partition another checkout’s directory owns, warning once per writer. That rule now has two implementations. OwnedGraphPartition.prepare (graphitron/src/main/java/no/sikt/graphitron/rewrite/diagnostics/OwnedGraphPartition.java) is the shared one the two diagnostics-stratum loaders call; CompileFacts.writeRound (graphitron/src/main/java/no/sikt/graphitron/rewrite/compile/CompileFacts.java) still carries the same logic inline, because the helper was extracted for the new loaders rather than by collapsing the site it was modelled on. The two agree today and the divergence is silent when they stop: the ownership refusal is the guard that keeps one module of a shared workspace from erasing another’s rows, so a rule change applied to one copy and not the other means one family respects a boundary its neighbours ignore, with no test failing.

The collapse is small: point writeRound at prepare, keep the javadoc reasoning about the minted anchor’s last_captured value with the helper (it is the same argument at every site), and widen the helper’s visibility past …rewrite.diagnostics, since CompileFacts lives in …rewrite.compile. Two shapes to settle while doing it. First, whether the helper should own the per-writer message tail as well: today each caller passes its own logger and warns with its own family named in the sentence ("that partition’s javac diagnostics", "that partition’s diagnostics"), which is worth keeping, so the parameter probably becomes the family’s display name rather than the whole message. Second, whether the warn-once cell belongs in the helper’s signature at all: CompileFacts uses a plain field, the new loaders a one-element array purely because the helper is static, and one shared instance-per-writer holder would read better than either. Adjacent: RejectionFacts and BuildWarningFacts each carry their own copy of the location-transcription block (the source name as read when it is non-empty, line and column when the line is positive), which is the same three lines and belongs at the same one site. It used to convert the source name to a URI on the way in and no longer does, so the block is smaller than it was when this item was filed, and the one site is a plainer transcription.