docs: changelog and blog post for v0.10.0

[nicoloboschi]

Changelog entry and release blog post for v0.10.0, prepared ahead of the release — the tag does not exist yet and no release was triggered.

generate-changelog refuses to run without the tag, so v0.10.0 was created locally on origin/main, used to generate the entry, and deleted immediately. It was never pushed, so release.sh can still create it normally.

Contents

• hindsight-docs/src/pages/changelog/index.md — generated 0.10.0 entry: 1 breaking change, 12 features, 7 improvements, 19 fixes, and one migration (e2f4a6c8b0d1, attachments + document_attachments — both new and empty, so it is quick on any deployment size).
• hindsight-docs/blog/2026-09-09-version-0-10-0.md — hand-written release notes.
• skills/hindsight-docs/references/ — changelog mirror, plus a pre-existing developer/extensions.md drift the skill regen picked up (the StaticKeys extension docs from feat(extensions): add StaticKeysTenantExtension — env-configured per-user API keys with per-schema isolation #3675 had not been mirrored).

What the blog post covers

• Images and files as first-class retain content (feat(retain): inline images and files as first-class content #4077)
• Throughput: pure-ASGI middleware (perf(api): replace both BaseHTTPMiddleware with pure ASGI (3.2x on cheap routes) #4235), concurrent embedding batches (perf(embeddings): issue a remote provider's batches concurrently (#4039) #4043), faster token counting (perf(tokenizer): replace tiktoken with quicktok and default to o200k_base #3788)
• Prompt preview for every operation + retain prompt tester (feat(api,control-plane): a prompt tester for retain, and prompt preview for every operation #4140)
• Fuzzy tag matching (feat(recall): fuzzy tag matching on tag_groups leaves (#4026) #4028) and multi-text entity labels (feat(entity-labels): add an open-vocabulary multi-valued label type (multi-text) #4027)
• Knowledge pages and mental models in document transfers (feat: export knowledge base in document transfers #3855)
• Extensions registry + Supabase unbundling (feat(extensions): add hindsight-extensions registry and unbundle Supabase #3988) and StaticKeysTenantExtension (feat(extensions): add StaticKeysTenantExtension — env-configured per-user API keys with per-schema isolation #3675), including the ModuleNotFoundError upgrade step for anyone on the built-in Supabase path
• Meta Model API provider (feat(llm): add Meta Model API as a first-class provider #4081), worker bank rotation (fix(worker): rotate slots across banks so bulk ingest stops starving them (#3861) #3980), mental-model list defaulting to metadata (feat(mental-models): default list to metadata; MCP list returns metadata only #4225)
• Breaking: bank profile and background endpoints now answer 410 Gone (feat(api)!: retire the bank profile and background endpoints #4127)

Rebased onto origin/main after #4234 reverted free-threaded CPython 3.14 (#4037, #4067) — both are out of the changelog and the blog post.

The per-bank text-search switch and the webhook signature work are in the changelog but not the blog post.

Second commit: a generator fix

The first pass produced 37 entries for 221 core commits, against 65 for the 145-commit release before it — whole user-facing fixes went unlisted. Three causes, measured separately on that range:

change	entries
baseline	37
max_completion_tokens 16000 → 100000	45
+ scoped grouping / "user-facing" rules	126
+ token-batched parallel summarization	152

• max_completion_tokens was a shared budget on a reasoning model — reasoning bills against it, so a long commit list spends most of it before the first entry.
• The prompt rules read as licence to merge unrelated fixes and to drop anything whose commit title sounds internal. Largest single effect.
• One LLM call for the whole release. Commits are now batched by token count and summarized concurrently, then deduped. Holding model and prompt fixed, one call gives 120 entries against 152 for five batches: 40 commits get an entry only when batched, 8 only when not.

The input was never the constraint. A 245-commit release is ~8500 tokens of commit subjects in total, so every arrangement fits the context window many times over — a batch budget near the context window would be a single batch that changes nothing. The budget is 2000 tokens; what it bounds is how many commits the model weighs at once before it starts summarizing them away.

Default model is now gpt-5.6-terra. tests/test_generate_changelog_batching.py covers the batching arithmetic and the dedup guarantees with the LLM stubbed.

Before merging

• main has moved on (fix(api): expose the bank template import request body #4247 landed after this branch was cut). The changelog is generated against the tag, so it will need a regenerate at release time whatever else happens — treat this entry as a draft of the shape, not the final list.

• No hero image. Recent release posts carry /img/blog/version-X-Y-Z-release.png; this one still needs one.

• openapi.json in the skill still reads 0.9.2 — correct today, but it needs a re-sync once release.sh bumps the version.