Add CSV table mode: single-relation loading without row ID

Introduces a `table` form for `csv_data` that loads all selected columns
into one output relation with schema `key=[T1..TN], value=[]`, dropping
the per-column row-ID keying of the existing `columns` form.

Proto:
- Add `CsvTarget` message (target_id, column_names, types)
- Add `optional CsvTarget target = 5` to `CSVData`

Grammar:
- Add `csv_table` rule: `(table <rel_id> ["col"...] [Type...])`
- Rewrite `csv_data` to accept either `gnf_columns?` or `csv_table?`
  in a single alternative (avoids LL(3) conflict on the shared prefix)
- Add construct/deconstruct helpers for the two modes

Regenerate parsers, pretty-printers, and proto stubs for Python,
Julia, and Go. Add `tests/lqp/csv_table.lqp` with four table-mode
variants and corresponding binary/snapshot artifacts.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Author: hbarthels

meta/src/meta/grammar.y

@@ -91,6 +91,7 @@
 %nonterm csv_data logic.CSVData
 %nonterm csv_locator_inline_data String
 %nonterm csv_locator_paths Sequence[String]
+%nonterm csv_table logic.CSVTarget
 %nonterm csvlocator logic.CSVLocator
 %nonterm data logic.Data
 %nonterm date logic.DateValue
@@ -1104,14 +1105,23 @@ gnf_columns
 csv_asof
     : "(" "asof" STRING ")"
 
+csv_table
+    : "(" "table" relation_id "[" STRING* "]" "[" type* "]" ")"
+      construct: $$ = logic.CSVTarget(target_id=$3, column_names=$5, types=$8)
+      deconstruct:
+        $3: logic.RelationId = $$.target_id
+        $5: Sequence[String] = $$.column_names
+        $8: Sequence[logic.Type] = $$.types
+
 csv_data
-    : "(" "csv_data" csvlocator csv_config gnf_columns csv_asof ")"
-      construct: $$ = logic.CSVData(locator=$3, config=$4, columns=$5, asof=$6)
+    : "(" "csv_data" csvlocator csv_config gnf_columns? csv_table? csv_asof ")"
+      construct: $$ = construct_csv_data($3, $4, $5, $6, $7)
       deconstruct:
         $3: logic.CSVLocator = $$.locator
         $4: logic.CSVConfig = $$.config
-        $5: Sequence[logic.GNFColumn] = $$.columns
-        $6: String = $$.asof
+        $5: Optional[Sequence[logic.GNFColumn]] = deconstruct_csv_data_columns_optional($$)
+        $6: Optional[logic.CSVTarget] = deconstruct_csv_data_target_optional($$)
+        $7: String = $$.asof
 
 csv_locator_paths
     : "(" "paths" STRING* ")"
@@ -1751,6 +1761,34 @@ def deconstruct_iceberg_data_to_snapshot_optional(msg: logic.IcebergData) -> Opt
     return builtin.none()
 
 
+def construct_csv_data(
+    locator: logic.CSVLocator,
+    config: logic.CSVConfig,
+    columns_opt: Optional[Sequence[logic.GNFColumn]],
+    target_opt: Optional[logic.CSVTarget],
+    asof: String,
+) -> logic.CSVData:
+    return logic.CSVData(
+        locator=locator,
+        config=config,
+        columns=builtin.unwrap_option_or(columns_opt, list[logic.GNFColumn]()),
+        asof=asof,
+        target=target_opt,
+    )
+
+
+def deconstruct_csv_data_columns_optional(msg: logic.CSVData) -> Optional[Sequence[logic.GNFColumn]]:
+    if not builtin.has_proto_field(msg, "target"):
+        return builtin.some(msg.columns)
+    return builtin.none()
+
+
+def deconstruct_csv_data_target_optional(msg: logic.CSVData) -> Optional[logic.CSVTarget]:
+    if builtin.has_proto_field(msg, "target"):
+        return builtin.some(builtin.unwrap_option(msg.target))
+    return builtin.none()
+
+
 def construct_export_iceberg_config_full(
     locator: logic.IcebergLocator,
     config: logic.IcebergCatalogConfig,

proto/relationalai/lqp/v1/logic.proto

@@ -276,11 +276,18 @@ message BeTreeLocator {
   int64 tree_height = 3;
 }
 
+message CSVTarget {
+  RelationId target_id = 1;          // Output relation path
+  repeated string column_names = 2; // CSV column names in order (no METADATA$ columns)
+  repeated Type types = 3;          // Column types in order (same length as column_names)
+}
+
 message CSVData {
   CSVLocator locator = 1;
   CSVConfig config = 2;
   repeated GNFColumn columns = 3;
   string asof = 4; // Blob storage timestamp for freshness requirements
+  optional CSVTarget target = 5; // If present, single-relation (table) mode; mutually exclusive with columns
 }
 
 message CSVLocator {