perf(localdb): dereg API, O(1) backlog depth, byte-budget batching, prepare-once apply (0.2.0)

WP3.3 of the ScadaBridge architectural-review remediation. Six changes, all
wire-compatible with 0.1.x peers — no proto change, no lib_schema_version bump. The
on-disk bookkeeping schema goes to v2 and is upgraded in place when a file is opened.

DeregisterReplicated(table). There was no way to stop capturing once a file had been
registered, so a node that does not replicate kept paying three triggers inside every
write transaction, appending to an oplog nothing would ever drain. Drops the triggers and
prunes the table's oplog + row-version rows in one transaction; idempotent, and
deliberately usable on a table this process never registered, since stale triggers left by
an EARLIER build are the primary case. Deregistration must be symmetric across a pair, and
already is enforced: the handshake compares registered-table digests fail-closed, so a node
that drops a table its peer still replicates stops syncing loudly rather than diverging.

RegisterReplicated(table, baselineExistingRows: true). Closes the late-opt-in hole: the
snapshot streamer pages from __localdb_row_version, which only triggers populate, so a node
that started replicating LATE never sent its pre-existing rows — silently, forever. Seeding
alone would not have fixed it, because a node whose oplog starts at seq 1 presents no gap at
the handshake and is told it needs no snapshot; the seed therefore also flags needs_snapshot,
and the existing resync machinery carries the rows. The floor is HLC 0 stamped with this
node's id: every genuine HLC is a UTC millisecond shifted left 16 bits and so is strictly
greater, which makes a baselined row lose every comparison against a real remote write and
win only where the peer has no version of that key at all. Two nodes baselining the same key
both hold 0 and resolve by the node-id tie-break — convergent, and nothing better is
available for two rows of unordered antiquity. Opt-in, so the documented CDC contract is
unchanged unless asked. Pruning the ledger on deregister is safe precisely because
re-registration can re-baseline; the two APIs are a pair.

O(1) backlog depth. GetOplogDepth ran COUNT(*) over the entire unacked backlog, once a
minute from maintenance AND on every metrics scrape — a full scan of a million rows exactly
when a node is already behind. Now MAX(seq) - MAX(last_acked_seq, MIN(seq) - 1), floored at
0: three b-tree seeks, measured at 58 VM steps against 100,020 for the count on a 50k
backlog. Exact, because seq is AUTOINCREMENT and every prune path removes a prefix (ack
pruning by watermark, the row cap by count, the age cap by an HLC cutoff that rises with
seq). The MIN clamp is what keeps it exact after a cap prune has moved the horizon above a
stalled watermark — without it the caps would re-flag on every tick forever.

Byte-budget batching. MaxBatchSize was the only cap on an outbound message, which made it a
proxy for a byte budget that consumers had to tune against their widest row: ScadaBridge
pinned it to 16 because 70 KB config_json times the 500 default is ~35 MB against a 4 MB
gRPC limit. MaxBatchBytes (default 2 MB) now bounds delta and snapshot messages by summed
serialized size, with the row count demoted to a secondary cap. A row larger than the budget
is sent alone rather than stalling the stream on it. Since more, smaller messages would
otherwise mean more apply transactions and more acks, inbound delta batches and acks ALREADY
queued are coalesced — one transaction and one ack per burst, bounded by the sender's own
caps, and a no-op under a trickle.

LwwApplier prepares each statement once. It went through the generic transaction façade,
which builds a fresh SqliteCommand per call: three SQL parses per applied row, on the path
that drains a reconnect backlog. Statements are now held for the batch and rebound.

Dropped __localdb_oplog_hlc. It indexed (table_name, pk_json, hlc), which no query has ever
used — every oplog access path is by seq or a full hlc-predicate delete — while costing a
b-tree insert inside every captured consumer write.

184/184 tests pass (was 149). New coverage: schema upgrade drops the index and preserves
identity/watermarks/data, deregistration idempotency and trigger absence and cross-process
cleanup, baseline seeding at the floor with LWW safety proved through the real applier,
depth asserted equal to the counting query across every prune scenario plus a constant-time
assertion over a 200k backlog, byte-budget chunking including the one-giant-row edge on both
the delta and snapshot paths, prepare-once statement counting, and coalescing that must not
swallow the message that ends its run.

Version 0.2.0.

ZB.MOM.WW.LocalDb/src/ZB.MOM.WW.LocalDb.Replication/ReplicationOptions.cs

@@ -14,9 +14,44 @@ public sealed class ReplicationOptions
     /// <summary>How often the change pump flushes pending oplog deltas to the peer.</summary>
     public TimeSpan FlushInterval { get; set; } = TimeSpan.FromMilliseconds(250);
 
-    /// <summary>Maximum oplog entries sent per outbound batch.</summary>
+    /// <summary>
+    /// Secondary, row-count cap on an outbound batch (and the snapshot streamer's DB page size).
+    /// <see cref="MaxBatchBytes"/> is the primary cap.
+    /// </summary>
+    /// <remarks>
+    /// Historically this was the ONLY cap, which made it a proxy for a byte budget that consumers
+    /// had to tune by hand against their widest row — ScadaBridge pinned it to 16 because a 70 KB
+    /// <c>config_json</c> times the 500 default is ~35 MB against a 4 MB gRPC limit. With
+    /// <see cref="MaxBatchBytes"/> doing that job directly, this is back to being what its name
+    /// says: a bound on how many rows one message may carry.
+    /// </remarks>
     public int MaxBatchSize { get; set; } = 500;
 
+    /// <summary>
+    /// Primary cap on an outbound batch: the summed serialized size of the entries in one
+    /// <c>DeltaBatch</c> / <c>SnapshotBatch</c> message, in bytes. Default 2 MB.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Sized at half the default 4 MB gRPC receive limit, so the message framing, the oneof wrapper
+    /// and any per-entry overhead the estimate does not model all fit inside the margin.
+    /// </para>
+    /// <para>
+    /// <b>A single row larger than the budget is still sent, alone.</b> Skipping it would stall the
+    /// stream permanently on one oversized row; sending it gives the transport the chance to accept
+    /// it (gRPC limits are configurable) and, if it refuses, a loud transport error naming the size
+    /// rather than silent non-convergence.
+    /// </para>
+    /// <para>
+    /// <b>Wire compatibility:</b> none required. Batch size is a sender-side choice the protocol
+    /// already leaves free — <c>DeltaBatch</c>/<c>SnapshotBatch</c> carry a repeated field with no
+    /// declared count — so a node on this version interoperates with a peer on any earlier one in
+    /// both directions, with no proto change and no <c>lib_schema_version</c> bump. (ScadaBridge
+    /// site pairs stop and start together regardless; this simply does not depend on it.)
+    /// </para>
+    /// </remarks>
+    public int MaxBatchBytes { get; set; } = 2 * 1024 * 1024;
+
     /// <summary>Backlog ceiling: exceeding it flags a snapshot resync and prunes the oldest oplog rows.</summary>
     public long MaxOplogRows { get; set; } = 1_000_000;