[BUG] gfql remote chain predicates fail when node uses col 'id'

[lmeyerov]

Describe the bug

g.chain_remote([n({'id': 0}), e_forward(), n()]) throws a 400 error

yet local g.chain([n({'id': 0}), e_forward(), n()]) works locally

To Reproduce

encountered on ldbc,

nodes_df = pd.read_csv(f'{data_dir}/SF-{scale_factor}/social-network-sf{scale_factor}-projected-fk-nodes.csv', low_memory=False)
edges_df = pd.read_csv(f'{data_dir}/SF-{scale_factor}/social-network-sf{scale_factor}-projected-fk-edges.csv')

See graphistry-dev: notebook/lab/workspaces/auto-m/tree/notebooks/pygraphistry-demos-tcook/gfql/tcook/LDBC_LSQB_GFQL_Queries_v2-leo.ipynb

Expected behavior

Both succeed

Actual behavior

Only local works, remote gives exn:

File /opt/conda/lib/python3.10/site-packages/graphistry/compute/chain_remote.py:91, in chain_remote_generic(self, chain, api_token, dataset_id, output_type, format, df_export_args, node_col_subset, edge_col_subset, engine, validate)
     84 headers = {
     85     "Authorization": f"Bearer {api_token}",
     86     "Content-Type": "application/json",
     87 }
     89 response = requests.post(url, headers=headers, json=request_body)
---> 91 response.raise_for_status()
     93 # deserialize based on output_type & format
     95 if self._edges is None or isinstance(self._edges, pd.DataFrame):

File /opt/conda/lib/python3.10/site-packages/requests/models.py:1021, in Response.raise_for_status(self)
   1016     http_error_msg = (
   1017         f"{self.status_code} Server Error: {reason} for url: {self.url}"
   1018     )
   1020 if http_error_msg:
-> 1021     raise HTTPError(http_error_msg, response=self)

HTTPError: 400 Client Error:  for url: http://nginx/api/v2/etl/datasets/65fd77ad689c45a999a7dfa3e49df08c/gfql/all

Screenshots

PyGraphistry 0.35.3 on 2.41.17 (graphistry-dev)