fix(player+lidarr): mini-player sync race; durable approve + dedup

player: setQueueFromTracks fast-starts a single source at player-index 0
while the full queue is broadcast, so the transient currentIndexStream→0
emission clobbered the correct mediaItem with queue.value[0] (the first
track). Mini bar / playlist marker pinned to the wrong track until a
later index event (~the "passive ~30s recovery"). Track a logical-index
base so the player→queue mapping stays correct during the fill window;
also fixes the latent forward-fill auto-advance off-by-base.

lidarr #50: approving no longer fails when Lidarr is down. Approve
records the decision durably first, then best-effort adds; the
reconciler idempotently (re)sends unconfirmed adds every tick until they
stick (new additive lidarr_add_confirmed_at; AddArtist/AddAlbum map
Lidarr's "already exists" 400 → ErrAlreadyExists). No failed-state or
expiry by design — Lidarr keeps trying, operator monitors.

lidarr #51: Create() is now idempotent — a non-terminal request for the
same MBID returns the existing row instead of inserting a duplicate.

Rewrites the obsolete LidarrUnreachable_503 test to assert the durable-
approve contract; threads a client factory into NewReconciler.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

internal/lidarr/client.go

@@ -4,6 +4,7 @@ import (
 	"bytes"
 	"context"
 	"encoding/json"
+	"errors"
 	"fmt"
 	"io"
 	"net/http"
@@ -300,12 +301,29 @@ func (c *Client) AddArtist(ctx context.Context, p AddArtistParams) error {
 	}
 	resp, err := c.post(ctx, "/api/v1/artist", body)
 	if err != nil {
+		if isAlreadyExists(err) {
+			return ErrAlreadyExists
+		}
 		return err
 	}
 	_ = resp.Body.Close()
 	return nil
 }
 
+// isAlreadyExists detects Lidarr's 400 "already in your library" rejection
+// from an add endpoint. The post() helper buckets it as ErrLookupFailed
+// but embeds Lidarr's errorMessage body snippet; matching the phrase lets
+// the idempotent retry path treat a re-add as success instead of error.
+func isAlreadyExists(err error) bool {
+	if !errors.Is(err, ErrLookupFailed) {
+		return false
+	}
+	m := strings.ToLower(err.Error())
+	return strings.Contains(m, "already exists") ||
+		strings.Contains(m, "already been added") ||
+		strings.Contains(m, "already in your library")
+}
+
 // AddAlbum posts to POST /api/v1/album. Returns nil on 2xx; typed error
 // otherwise.
 func (c *Client) AddAlbum(ctx context.Context, p AddAlbumParams) error {
@@ -324,6 +342,9 @@ func (c *Client) AddAlbum(ctx context.Context, p AddAlbumParams) error {
 	}
 	resp, err := c.post(ctx, "/api/v1/album", body)
 	if err != nil {
+		if isAlreadyExists(err) {
+			return ErrAlreadyExists
+		}
 		return err
 	}
 	_ = resp.Body.Close()

internal/lidarr/errors.go

@@ -14,6 +14,12 @@ var (
 	ErrLookupFailed   = errors.New("lidarr: lookup failed") // 4xx other than 401/403
 	ErrServerError    = errors.New("lidarr: server error")  // 5xx
 	ErrInvalidPayload = errors.New("lidarr: invalid payload")
+	// ErrAlreadyExists is returned by AddArtist/AddAlbum when Lidarr
+	// rejects the add with a 400 because the artist/album is already in
+	// its library. This is the desired end state, not a failure — the
+	// idempotent retry path (reconciler re-sending an add after a partial
+	// success) treats it as confirmed rather than spinning forever.
+	ErrAlreadyExists = errors.New("lidarr: already exists")
 	// ErrNotFound is returned by LookupArtistByMBID and LookupAlbumByMBID
 	// when Lidarr returns 200 with an empty array — i.e., the MBID isn't in
 	// Lidarr's monitored set. Distinguished from network/auth errors so admin