Add an Objective registry for the flow/ddpm/wgan generator choice (gitea #32)

generator ∈ {"flow", "ddpm", "wgan"} was tested as a bare string in ~45
sites across models.py, sample.py, builders.py, trainers.py, and
stage2_inputs.py, each independently re-deriving one of five consequences
of the choice (needs a time embedding? what does the trunk take as input?
is the type slice folded into the trunk output? which sampler? which
loss?). giant/model/objectives.py adds an Objective ABC + OBJECTIVE_REGISTRY
+ build_objective factory, mirroring routers.py's Router pattern, and every
bare-string site now goes through it (needs_time, is_adversarial,
folds_type_slice, trunk_in_dim, build_schedule, stage1_loss/stage2_loss).

Per discussion: FlowDDPMStageTrainer and WGANStageTrainer stay separate
classes rather than merging into one StageTrainer as the issue's sketch
proposed — their training loops are genuinely different shapes (single loss
vs. dual G/D step with gradient penalty/n_critic/ST-Gumbel), and trainers.py
is the least-covered-by-fast-tests part of the codebase, so a full merge
was judged out of proportion to this issue's risk budget.
FlowDDPMStageTrainer's own loss dispatch (flow vs ddpm, one-shot vs AR) does
move onto the objective, so a future non-adversarial objective (rectified
flow, consistency distillation) is still a one-file, zero-trainer-edits
addition.

No config-schema change — stage{1,2}_model.generator stays the persisted
string, just looked up in the registry instead of string-compared. An
unrecognized generator value now fails fast with a clear ValueError instead
of silently falling through some bare-string checks and not others (same
behavior build_router/build_trunk already have for their own type keys).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

giant/sample.py

@@ -2,7 +2,7 @@ import torch
 import torch.nn.functional as F
 
 from giant.constants import CONT_SLOT_DIM, X_DIM
-from giant.model.network import Stage2Autoregressive, stage2_trunk_sec_dim
+from giant.model.network import DdpmObjective, Stage2Autoregressive, build_objective, stage2_trunk_sec_dim
 from giant.model.schedule import CosineSchedule
 
 
@@ -132,7 +132,7 @@ def _stage2_flat_width(sec_decoder: torch.nn.Module) -> int:
 
 def _type_folded(sec_decoder: torch.nn.Module) -> bool:
     target = sec_decoder.particle_type_cfg.get("target", "physical")
-    return target == "physical" or sec_decoder.generator_kind == "wgan"
+    return target == "physical" or build_objective(sec_decoder.generator_kind).folds_type_slice
 
 
 def _decode_stage2_flat(
@@ -281,7 +281,7 @@ def sample_secondaries_ar(
     device = cond_cont.device
     k_max = sec_decoder.k_max
     type_dim = sec_decoder.type_dim
-    generator = sec_decoder.generator_kind
+    objective = build_objective(sec_decoder.generator_kind)
     target = sec_decoder.particle_type_cfg.get("target", "physical")
     type_folded = _type_folded(sec_decoder)
     token_dim = CONT_SLOT_DIM + type_dim if type_folded else CONT_SLOT_DIM
@@ -301,7 +301,7 @@ def sample_secondaries_ar(
         slot_idx = torch.full((B, 1), k / max(k_max - 1, 1), device=device, dtype=torch.float32)
         hist, history_cache = sec_decoder.history_step(history_feat, has_prev, history_cache)
 
-        if generator == "wgan":
+        if objective.is_adversarial:
             z = torch.randn(B, 1, sec_decoder.noise_dim, device=device)
             token = sec_decoder(
                 z,
@@ -383,10 +383,10 @@ def sample_stage1(
     ddpm_steps: int = 1000,
 ) -> tuple[torch.Tensor, torch.Tensor | None]:
     """Dispatches on `stage1_model.generator_kind`."""
-    kind = stage1_model.generator_kind
-    if kind == "wgan":
+    objective = build_objective(stage1_model.generator_kind)
+    if objective.is_adversarial:
         return sample_wgan(stage1_model, cond_cont, cond_cat)
-    if kind == "ddpm":
+    if isinstance(objective, DdpmObjective):
         schedule = CosineSchedule(T=ddpm_steps).to(cond_cont.device)
         return sample_ddpm(stage1_model, cond_cont, cond_cat, schedule)
     return sample_flow(stage1_model, cond_cont, cond_cat, steps=steps)
@@ -409,7 +409,7 @@ def sample_stage2(
     """
     if isinstance(sec_decoder, Stage2Autoregressive):
         return sample_secondaries_ar(sec_decoder, cond_cont, cond_cat, stage1_out, n_sec_pred, steps=steps)
-    if sec_decoder.generator_kind == "wgan":
+    if build_objective(sec_decoder.generator_kind).is_adversarial:
         return sample_secondaries_wgan(sec_decoder, cond_cont, cond_cat, stage1_out, n_sec_pred)
     return sample_secondaries(sec_decoder, cond_cont, cond_cat, stage1_out, n_sec_pred, steps=steps)