feat(gfql): close encoding call parity gaps (encode_edge_icon / encode_axis) + strengthen encode validators

[lmeyerov]

Problem

GFQL call validation has parity and depth gaps for encoding operations used in downstream remote flows.

Findings (audit date: 2026-04-29)

1) Missing encoding operations in call safelist

In graphistry/compute/gfql/call/validation.py safelist entries include:

• encode_point_color
• encode_edge_color
• encode_point_size
• encode_point_icon

But no safelist entries for:

• encode_edge_icon
• encode_axis

At the same time, Plottable exposes both methods (encode_edge_icon, encode_axis).

2) Typing contract mirrors the same omission

graphistry/models/gfql/types/call.py CallMethodName also excludes encode_edge_icon and encode_axis, reinforcing runtime/type-level mismatch.

3) Existing encoding validation is shallow

For present encode ops, validators often only check container type (e.g., categorical_mapping: is_dict, palette: list) without key/value semantic checks.

Why this matters

• Downstream tools using GFQL calls for remote execution get inconsistent capabilities vs direct Plottable APIs.
• LLM/tooling flows need clear, early, actionable validation errors for encoding payload mistakes.
• Current mismatch can push users toward ad-hoc settings-level workarounds.

Requested behavior

1. Decide and document parity policy:

• either add encode_edge_icon and encode_axis to GFQL call safelist + typed call model, or
• explicitly document they are intentionally unsupported in call mode.

2. If supported, add validators + executor support + tests for both ops.

3. Improve encode validator depth for common payloads:

• categorical_mapping key/value type checks
• palette entry type checks
• clear per-field error messages

Related

• Deep ring axis validator request: feat(gfql): deep axis payload validation for ring_*_layout calls #1239
• Strict reusable settings validator request: feat(validation): expose strict reusable settings validators #1240
• graphistry-js encodeAxis contract issue: feat(client-api): harden encodeAxis input validation + schema clarity graphistry-js#195