feat(run): durable checkpoint + resume (wire Ports.Checkpointer)

The kernel defined run.Ports.Checkpointer + the checkpoint battery but never
drove them (the documented "P2 follow-up"). This wires durable recovery into
the run loop so a run interrupted by shutdown can resume on the next boot
instead of being lost — the executus-side half of mort's durable-agent-recovery
parity (mort #1355).

Kernel (run/):
- Ports.Checkpointer is now a CheckpointerFactory (Begin per run → a per-run
  Checkpointer, or nil for a non-durable run). The single per-instance
  Checkpointer couldn't distinguish runs; a factory mints one per run, matching
  mort's agentexec.CheckpointerFactory.
- RunInfo gains GuildID + ModelTier (so the factory can build resume meta);
  RunCheckpointState gains CompletedPhases + ActivePhase (+ PhaseOutput).
- run/checkpoint.go: ResumeState + WithResumeState / WithExistingCheckpointer
  context carriers, classifyCheckpointOutcome (success→Complete, shutdown→leave
  for boot recovery, else→Fail using run.ErrShutdown), and finalizeCheckpoint.
- run/executor.go: resolve the per-run checkpointer (existing-from-ctx on a
  recovery re-run, else factory.Begin); single-loop wraps the step observer to
  accumulate the transcript + Save each step (host throttles), and a recovered
  run seeds the saved transcript via WithHistory and continues with no new
  input; finalize on exit.
- run/phases.go: phase-boundary checkpointing — record completed phases after
  each phase; a resumed run skips already-completed phases (the interrupted
  phase re-runs from its start — boundary-granular, documented; only the
  single-loop path resumes mid-loop).

Battery (checkpoint/): NewFactory wires the battery into the factory port
(per-run handle, meta derived from RunInfo); RunCheckpoint + handle.Save carry
the phase fields.

Tests (run/checkpoint_test.go): the finalize decision matrix; single-loop
Save+Complete; terminal-error Fail; resume seeds history; phase-boundary Saves
completed phases; resume skips completed phases. Full ./... green.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

run/ports.go

@@ -33,9 +33,10 @@ type Ports struct {
 	Budget Budget
 	// Critic optionally monitors a long run for hangs/runaways. nil = none.
 	Critic Critic
-	// Checkpointer persists resumable progress for durable recovery. nil = no
-	// checkpointing (a run interrupted by shutdown is simply lost).
-	Checkpointer Checkpointer
+	// Checkpointer mints a per-run Checkpointer for durable recovery (it decides
+	// per run whether the run is durable). nil = no checkpointing (a run
+	// interrupted by shutdown is simply lost).
+	Checkpointer CheckpointerFactory
 	// Palette resolves SkillPalette / SubAgentPalette entries into delegation
 	// tools (skill__<name> / agent__<name>). nil = those entries are inert.
 	Palette PaletteSource
@@ -66,7 +67,9 @@ type RunInfo struct {
 	Name        string
 	CallerID    string
 	ChannelID   string
+	GuildID     string // the originating guild/server id (empty for DMs/triggers)
 	ParentRunID string
+	ModelTier   string // the run's resolved base tier (for checkpoint re-dispatch)
 	Inputs      map[string]any
 	StartedAt   time.Time
 	// MaxIterations is the run's base tool-dispatch step ceiling, so a critic can
@@ -172,6 +175,16 @@ type CriticHandle interface {
 
 // --- Checkpointer ---
 
+// CheckpointerFactory decides, per run, whether the run is durable and (if so)
+// mints the per-run Checkpointer that records its progress. It returns (nil, nil)
+// for a non-durable run (the common short-run case — no checkpointing overhead).
+// A storage error should be logged and degraded to (nil, nil) so a failing
+// checkpoint store never fails the run. Mirrors mort's
+// agentexec.CheckpointerFactory.
+type CheckpointerFactory interface {
+	Begin(ctx context.Context, info RunInfo) (Checkpointer, error)
+}
+
 // Checkpointer persists a run's resumable progress for durable recovery.
 // Mirrors mort's agentexec.RunCheckpointer.
 type Checkpointer interface {
@@ -184,11 +197,24 @@ type Checkpointer interface {
 	Fail(ctx context.Context, err error) error
 }
 
-// RunCheckpointState is the resumable snapshot a Checkpointer persists. Kept
-// minimal here; the executor extends what it records during the merge.
+// RunCheckpointState is the resumable snapshot a Checkpointer persists.
 type RunCheckpointState struct {
+	// Messages is the running transcript (single-loop run) OR the active phase's
+	// transcript (multi-phase run). May be nil.
 	Messages  []llm.Message
 	Iteration int
+	// CompletedPhases is set only for multi-phase runs: the outputs of phases
+	// already finished, in phase order. nil for single-loop runs.
+	CompletedPhases []PhaseOutput
+	// ActivePhase is the name of the in-progress phase (multi-phase only).
+	ActivePhase string
+}
+
+// PhaseOutput is one completed pipeline phase's name and output text, recorded in
+// a checkpoint so a resumed multi-phase run can skip already-finished phases.
+type PhaseOutput struct {
+	Name   string
+	Output string
 }
 
 // --- PaletteSource ---