feat(core): add first-class Pi session indexing (#23)

Pi cannot be read as another linear JSONL stream. Its history is a tree with
a durable leaf, orphan roots, branch summaries, and two compaction forms, so
the active context is something the format states rather than something line
order implies. The adapter keeps those semantics inside itself and projects
the result into the existing canonical tables.

Sessions are keyed by (normalized header cwd, header id) rather than by path,
because Pi's --session-id lookup is project-local: two projects may reuse an
id, while a move or an identical copy is still one session. Discovery covers
both layouts Pi writes and fingerprints each file by mtime, ctime, size and
inode, so a rewrite that preserves mtime is not read as unchanged.

Abandoned branches are preserved rather than dropped. Visibility becomes
three-state -- visible, inactive, hidden -- and helpers return only visible
rows until includeInactive asks for the superseded path, labeling every row
so a caller knows which it holds. Usage counts all three, because an
abandoned call still spent tokens; message_count reports only the visible
transcript.

A committed MIT-licensed oracle transcribed from Pi 0.83.0 pins the context
algorithms, and a fixed-seed differential runs 512 generated sessions against
it on every test run. Schema changes are additive.

skill-doc/references/retrieval-semantics.md

@@ -37,7 +37,7 @@ Project-like fields are distinct:
 - `memories.project`: stored project slug copied onto registered memory records.
 - `sessions.project_path`: absolute session path derived from message `cwd` when available; slug decoding is only a fallback.
 - `messages.cwd`: working directory at message time.
-- `sessions.source` / `messages.source`: transcript provider, currently `claude` or `codex`.
+- `sessions.source` / `messages.source`: transcript provider: `claude`, `codex`, `kimi`, or `pi`.
 - helper `project`: SQL `LIKE` over `sessions.project`, not exact membership.
 - helper `source`: optional provider filter. Omit it unless provenance matters.
 
@@ -85,6 +85,9 @@ Ordering and context are semantic:
 - `fileHistory()` is oldest first.
 - `search().context` is temporal neighbors in one session, not causal context.
 - `context(uuid)` and `trace(uuid)` are for parent-chain/causal expansion.
+  They return only current evidence by default. Use `includeInactive: true` for
+  a Pi path that was tried and then superseded; hidden records remain
+  unavailable.
 
 ### Evidence Before Conclusion