Adopt drift-aware publishing: pubmate v0.2.2 + updated assertion template

[tkuhn]

Upgrades the publishing toolchain to the drift-aware pubmate release and points the workflow at the updated "Defining a biochementity" assertion template.

Changes

• pyproject.toml — pin pubmate v0.2.1 → v0.2.2. v0.2.2 lands eu-parc/pubmate#8: publish_incremental() / fingerprint_term() detect content or wrapper drift (template, part-of, nanopub-type, license) on already-published terms and supersede them in place — keeping thing URIs stable and resolving inter-term links — instead of the old append-only behaviour. migrate now records a drift fingerprint per term.
• Makefile — NANOPUB_TEMPLATE → the updated template https://w3id.org/np/RALm3XedpEbjtQy1nPJEQpOdsV0hPm5APkvvym-7P1Vpk (was RAhSlIuuw5…).

Behavioural note — this triggers a mass supersede

The currently-committed published/*.trig were minted with the old template. Because the template is part of the drift fingerprint, the next make migrate / make publish-defining run will see wrapper-drift on all 882 terms and supersede every one (new nanopub versions, npx:supersedes chain, thing URIs unchanged). This is intended — it is exactly what drift-aware publishing is for — but it is a deliberate, network-visible re-issue, so merge when you're ready to run that.

Verification

• pubmate v0.2.2 includes the add identifiers #8 drift merge; verified from pubmate import publish_incremental, fingerprint_term, FP_SCHEME imports from the pinned build.
• Full make migrate DRY=--dry-run on the pinned toolchain: 882 defining nanopubs minted, 21 cyclic supersessions, 0 dangling references, and the id-map is written in the new 4-column schema with a fingerprint on every term.

🤖 Generated with Claude Code

[tkuhn]

I updated the template to cover all biochementity nanopubs as we are currently creating them. This PR uses the latest pubmate version, and sets the template field so the next update will re-publish all of them with a fully compliant template reference.