fix(weave): shard call_parts by id so call_start/call_end co-locate

call_parts was using the default rand() sharding key, so call_start and
call_end for the same call could land on different shards. Once split,
the partial-state rows can never merge in calls_merged_local (OPTIMIZE
runs per-shard), and queries that filter on an aggregated column see
inconsistent state.

Concretely: call_end doesn't carry parent_id, so its row defaults to
NULL. Filters like trace_roots_only (`parent_id IS NULL`) then match
the call_end row of every child call as if it were a root, inflating
counts.

Shard by `id` instead of `wf_clickhouse_calls_shard_key()` (which
defaults to trace_id): trace_id is Nullable on call_end so sipHash64
returns Nullable, which ClickHouse rejects as a sharding expression
(TYPE_MISMATCH). `id` is non-null on every call_part row and uniquely
identifies a call, so all parts of one call land together.

calls_merged Distributed table is intentionally left rand(): the only
writes come from the MV which fires on the local source/target pair and
never goes through the Distributed wrapper.

Author: gtarpenning

weave/trace_server/clickhouse_trace_server_migrator.py

@@ -162,8 +162,13 @@ def _is_transient_ch_error(exc: BaseException) -> bool:
 # Valid values: "trace_id" (default), "id", "project_id"
 ID_SHARDED_TABLES: dict[str, str] = {
     "calls_complete": wf_clickhouse_calls_shard_key(),
-    # Keep one trace/turn on one shard. This matches the default calls sharding
-    # key and keeps trace detail reads local.
+    # call_parts (call_start/call_end rows) must shard by `id`, not `trace_id`:
+    # call_end rows don't carry trace_id (only call_start does), so sharding by
+    # trace_id sends them via rand() and they land on a different shard than
+    # the matching call_start. Without co-location the partial states never
+    # merge, and queries that filter on aggregated columns (e.g. parent_id IS
+    # NULL for trace_roots_only) match the call_end row of every child call.
+    "call_parts": "id",
     "spans": "trace_id",
     "messages": "trace_id",
     # Keep each agent aggregate on one shard. Shard versions by the same key so