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rows until includeInactive asks for the superseded path, labeling every row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.

packages/core/src/parsing.ts

@@ -6,6 +6,8 @@ import { closeSync, existsSync, openSync, readSync, readdirSync, statSync } from
 import { homedir } from 'node:os';
 import { isAbsolute, join, normalize } from 'node:path';
 
+import type { InventoryIssue } from './providers/types.ts';
+
 const CLAUDE_DIR = join(homedir(), '.claude');
 const CODEX_DIR = join(homedir(), '.codex');
 const PROJECTS_DIR = join(CLAUDE_DIR, 'projects');
@@ -15,6 +17,12 @@ const TEXT_LIMIT = 10000;
 type JsonRecord = Record<string, any>;
 type JsonValue = any;
 
+function sourceInventoryIssue(path: string, error: unknown): InventoryIssue {
+  return { path, error: error instanceof Error ? error.message : String(error) };
+}
+
+type DiscoveryIssueHandler = (issue: InventoryIssue) => void;
+
 export interface ClaudeJsonlFile {
   path: string;
   sessionId: string;
@@ -153,16 +161,19 @@ function inferProjectPath(project: string | null | undefined, observedCwds: unkn
   return best?.path || legacyProjectPathFromSlug(project);
 }
 
-function discoverJsonlFiles(projectsDir = PROJECTS_DIR): ClaudeJsonlFile[] {
+function discoverJsonlFiles(
+  projectsDir = PROJECTS_DIR,
+  reportIssue?: DiscoveryIssueHandler,
+): ClaudeJsonlFile[] {
   const files: ClaudeJsonlFile[] = [];
   if (!existsSync(projectsDir)) return files;
   let projects;
-  try { projects = readdirSync(projectsDir); } catch (e) { process.stderr.write(`Warning: cannot read projects dir: ${e instanceof Error ? e.message : String(e)}\n`); return files; }
+  try { projects = readdirSync(projectsDir); } catch (error) { reportIssue?.(sourceInventoryIssue(projectsDir, error)); return files; }
   for (const proj of projects) {
     const projPath = join(projectsDir, proj);
     if (!isDir(projPath)) continue;
     let entries;
-    try { entries = readdirSync(projPath); } catch { continue; }
+    try { entries = readdirSync(projPath); } catch (error) { reportIssue?.(sourceInventoryIssue(projPath, error)); continue; }
     for (const f of entries) {
       if (f.endsWith('.jsonl'))
         files.push({ path: join(projPath, f), sessionId: f.slice(0, -6), project: proj, isSubagent: false });
@@ -171,7 +182,7 @@ function discoverJsonlFiles(projectsDir = PROJECTS_DIR): ClaudeJsonlFile[] {
       const saDir = join(projPath, sd, 'subagents');
       if (!isDir(saDir)) continue;
       let saEntries;
-      try { saEntries = readdirSync(saDir); } catch { continue; }
+      try { saEntries = readdirSync(saDir); } catch (error) { reportIssue?.(sourceInventoryIssue(saDir, error)); continue; }
       for (const sf of saEntries) {
         if (sf.endsWith('.jsonl'))
           files.push({ path: join(saDir, sf), sessionId: sd, project: proj, isSubagent: true, agentId: sf.slice(0, -6) });
@@ -179,12 +190,12 @@ function discoverJsonlFiles(projectsDir = PROJECTS_DIR): ClaudeJsonlFile[] {
       const wfRoot = join(saDir, 'workflows');
       if (!isDir(wfRoot)) continue;
       let wfDirs;
-      try { wfDirs = readdirSync(wfRoot); } catch { continue; }
+      try { wfDirs = readdirSync(wfRoot); } catch (error) { reportIssue?.(sourceInventoryIssue(wfRoot, error)); continue; }
       for (const wfDir of wfDirs) {
         const wfPath = join(wfRoot, wfDir);
         if (!isDir(wfPath)) continue;
         let wfEntries;
-        try { wfEntries = readdirSync(wfPath); } catch { continue; }
+        try { wfEntries = readdirSync(wfPath); } catch (error) { reportIssue?.(sourceInventoryIssue(wfPath, error)); continue; }
         for (const wf of wfEntries) {
           if (wf.endsWith('.jsonl'))
             files.push({ path: join(wfPath, wf), sessionId: sd, project: proj, isSubagent: true, agentId: wf.slice(0, -6), workflowRunId: wfDir });
@@ -195,12 +206,15 @@ function discoverJsonlFiles(projectsDir = PROJECTS_DIR): ClaudeJsonlFile[] {
   return files;
 }
 
-function discoverCodexJsonlFiles(sessionsDir = CODEX_SESSIONS_DIR): CodexJsonlFile[] {
+function discoverCodexJsonlFiles(
+  sessionsDir = CODEX_SESSIONS_DIR,
+  reportIssue?: DiscoveryIssueHandler,
+): CodexJsonlFile[] {
   const files: CodexJsonlFile[] = [];
   if (!existsSync(sessionsDir)) return files;
   const walk = (dir: string): void => {
     let entries;
-    try { entries = readdirSync(dir, { withFileTypes: true }); } catch { return; }
+    try { entries = readdirSync(dir, { withFileTypes: true }); } catch (error) { reportIssue?.(sourceInventoryIssue(dir, error)); return; }
     for (const entry of entries) {
       const fp = join(dir, entry.name);
       if (entry.isDirectory()) {
@@ -369,7 +383,7 @@ export {
   CLAUDE_DIR, CODEX_DIR, PROJECTS_DIR, CODEX_SESSIONS_DIR, TEXT_LIMIT,
   trunc, truncJson, extractText, extractContentType, extractMessageIsMeta, isSkillInstructions, filePath, isDir, readLines,
   legacyProjectPathFromSlug, normalizeObservedCwd, projectSlugFromPath, inferProjectPath,
-  discoverJsonlFiles, discoverCodexJsonlFiles,
+  discoverJsonlFiles, discoverCodexJsonlFiles, sourceInventoryIssue,
   codexDbId, codexRawId, codexLineUuid, codexCallId, codexParentThreadId, codexIsGuardianThread,
   readCodexGuardianThreadInfo, codexAgentNickname, codexAgentRole, parseCodexJsonInput,
   codexUsage, codexEventText, codexMessagePayloadText, codexVisibleMessageKey, codexToolInput, codexToolOutput,