feat: harden storage, schema persistence, fair benches, fix wire crash

Core storage: hash MemTable, WAL group commit, L0 compaction rebuild,
reader-writer lock, and a global StorageGate so HTTP workers and TCP
share the LSM safely under multi-thread access.

Schema: durable CREATE/ALTER/DROP under _schema:tables:* with full LSM
restore on open. Executor types/values/schema split into query/exec/.

Wire protocol: switch default MM to ARC — ORC cycle collector segfaulted
after ~20 async INSERTs. Fair multi-tier benchmarks (SQLite/HTTP/wire/PG)
and honesty docs for mixed-tier comparisons.

nim.cfg

@@ -1,3 +1,8 @@
 -d:ssl
 --threads:on
 --path:"src"
+# ARC: ORC cycle collector crashes under async wire-protocol load
+# (markGray/trace SIGSEGV after ~20 sequential INSERTs). ARC is stable
+# for the TCP server + HTTP worker mix. Prefer breaking cycles over
+# re-enabling ORC without a reproducer.
+--mm:arc