Add GIMM-VFI support (NeurIPS 2024) with single-pass arbitrary-timestep interpolation

Integrates GIMM-VFI alongside existing BIM/EMA/SGM models. Key feature: generates
all intermediate frames in one forward pass (no recursive 2x passes needed for 4x/8x).

- Vendor gimm_vfi_arch/ from kijai/ComfyUI-GIMM-VFI with device fixes
- Two variants: RAFT-based (~80MB) and FlowFormer-based (~123MB)
- Auto-download checkpoints from HuggingFace (Kijai/GIMM-VFI_safetensors)
- Three new nodes: Load GIMM-VFI Model, GIMM-VFI Interpolate, GIMM-VFI Segment Interpolate
- single_pass toggle: True=arbitrary timestep (default), False=recursive like other models
- ds_factor parameter for high-res input downscaling

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

gimm_vfi_arch/generalizable_INR/modules/module_config.py

@@ -0,0 +1,52 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+# --------------------------------------------------------
+# References:
+# ginr-ipc: https://github.com/kakaobrain/ginr-ipc
+# --------------------------------------------------------
+
+from typing import List, Optional
+from dataclasses import dataclass, field
+from omegaconf import MISSING
+
+
+@dataclass
+class HypoNetActivationConfig:
+    type: str = "relu"
+    siren_w0: Optional[float] = 30.0
+
+
+@dataclass
+class HypoNetInitConfig:
+    weight_init_type: Optional[str] = "kaiming_uniform"
+    bias_init_type: Optional[str] = "zero"
+
+
+@dataclass
+class HypoNetConfig:
+    type: str = "mlp"
+    n_layer: int = 5
+    hidden_dim: List[int] = MISSING
+    use_bias: bool = True
+    input_dim: int = 2
+    output_dim: int = 3
+    output_bias: float = 0.5
+    activation: HypoNetActivationConfig = field(default_factory=HypoNetActivationConfig)
+    initialization: HypoNetInitConfig = field(default_factory=HypoNetInitConfig)
+
+    normalize_weight: bool = True
+    linear_interpo: bool = False
+
+
+@dataclass
+class CoordSamplerConfig:
+    data_type: str = "image"
+    t_coord_only: bool = False
+    coord_range: List[float] = MISSING
+    time_range: List[float] = MISSING
+    train_strategy: Optional[str] = MISSING
+    val_strategy: Optional[str] = MISSING
+    patch_size: Optional[int] = MISSING