refactor(groot): N1.7 style cleanup (utils, imports, flash-attn, config)

Mechanical refactor of the GR00T N1.7 policy to match the repo's architecture and
style standards. No change to policy algorithm/numerics; only UX/CLI and packaging
changes. Tests are intentionally left untouched (out of scope) and need updating
for the removed `model_version` field.

Cleanup & consolidation:
- Add `groot/utils.py` holding the pure, side-effect-free helpers (JSON I/O, value
  coercion, stat flattening, rot6d/SE3 math, language/batch prep) shared by the
  config and processor layers.
- Remove dead code: the unused `resolve_groot_n1_7_backbone_model` cache-resolver
  cluster, `GR00TN17Config.to_filtered_dict/json`, and the `_copy_default` wrapper.

Imports & execution guards:
- Hoist nested imports to module top; relative imports within the package, absolute
  for external modules. The version-gated Qwen3-VL classes import under the single
  `_transformers_available` guard (transformers is pinned >=5.4, which ships them).
- No import-time side effects: `_register_with_transformers()` now runs in
  `GR00TN17.__init__` (idempotent via `register(exist_ok=True)`), and the N1.5 step
  stubs register lazily before pipeline deserialization (idempotent via the
  registry, no run-once globals).
- Gate optional deps at the point of use with `require_package(..., extra="groot")`.

Dependencies & docs:
- Drop `flash-attn` (and its build-only dep `ninja`) from the `groot` extra; default
  to SDPA (numerically equivalent) with opt-in via `--policy.use_flash_attention`.
  Un-comment `lerobot[groot]` in the `all` extra and regenerate `uv.lock`.
- Rewrite the `groot.mdx` install section: flash-attn is a purely optional,
  user-managed optimization that LeRobot neither installs nor requires.

Config & CLI:
- Surface previously-frozen knobs on `GrootConfig` (plumbed into `GR00TN17Config`;
  no-ops at their defaults): inference — `num_inference_timesteps`, `rtc_ramp_rate`,
  `use_flash_attention`; fine-tuning — `tune_top_llm_layers` (partial-LLM tuning)
  and `tune_vlln` (previously hardwired to True).
- Convert the single-valued `model_version` and `n1_7_backbone_model` fields to
  internal constants.
- Keep `base_model_path`: it is NOT equivalent to `pretrained_path` (raw NVIDIA
  checkpoints have no LeRobot `type` field and load only via `base_model_path`) and
  is genuinely user-tunable.
- Keep the deprecated Isaac-GR00T/N1.5 fields (and the dead LoRA fields) as a
  back-compat block so a v0.5.1 N1.5 `config.json` still parses under draccus and is
  rejected with the friendly N1.5 removal message instead of an opaque decode error.

pyproject.toml

@@ -221,8 +221,6 @@ groot = [
     "dm-tree>=0.1.8,<1.0.0",
     "timm>=1.0.0,<1.1.0",
     "decord>=0.6.0,<1.0.0; (platform_machine == 'AMD64' or platform_machine == 'x86_64')",
-    "ninja>=1.11.1,<2.0.0",
-    "flash-attn>=2.5.9,<3.0.0 ; sys_platform != 'darwin'"
 ]
 sarm = ["lerobot[transformers-dep]", "pydantic>=2.0.0,<3.0.0", "faker>=33.0.0,<35.0.0", "lerobot[matplotlib-dep]", "lerobot[qwen-vl-utils-dep]"]
 robometer = ["lerobot[transformers-dep]", "lerobot[qwen-vl-utils-dep]", "lerobot[peft-dep]"]
@@ -308,7 +306,7 @@ all = [
     "lerobot[pi]",
     "lerobot[molmoact2]",
     "lerobot[smolvla]",
-    # "lerobot[groot]", TODO(Steven): Gr00t requires specific installation instructions for flash-attn
+    "lerobot[groot]",
     "lerobot[xvla]",
     "lerobot[hilserl]",
     "lerobot[vla_jepa]",