feat(adr-185): P2 MERIDIAN bindings (wifi_densepose.meridian) + parity harness

Bind the ADR-027 MERIDIAN cross-environment domain-generalization surface
into the wheel behind a gated [meridian] extra / Cargo `meridian` feature.
Inference/adaptation path only (tch-free), per ADR-185 section 3.3.

Surface (bound against the REAL code at HEAD, not the ADR wishlist):
- HardwareType / HardwareNormalizer / CanonicalCsiFrame (from
  wifi-densepose-signal::hardware_norm)
- MeridianGeometryConfig / GeometryEncoder (64-dim, permutation-invariant)
- RapidAdaptation / AdaptationResult (push_frame + adapt, LoRA deltas)
- CrossDomainEvaluator + mpjpe (from wifi-densepose-train, NO tch-backend)
All compute paths GIL-released (py.allow_threads).

Honest deviations from ADR section 3.3 (documented in the module header):
- ADR's RapidAdaptation.calibrate(windows) and AdaptationResult.converged
  DO NOT EXIST. Real API is push_frame + adapt(); result carries
  {lora_weights, final_loss, frames_used, adaptation_epochs}. Bound as-is.
- ADR's HardwareType.detect exposed as a staticmethod delegating to the
  real HardwareNormalizer::detect_hardware.
- ADR's normalize(frame: CsiFrame, hw) is really normalize(amplitude,
  phase, hw) over f64 vectors returning Result; bound faithfully.
- CanonicalCsiFrame fields are singular amplitude/phase (ADR said plural).
Training-time types (DomainFactorizer, GradientReversalLayer,
VirtualDomainAugmentor) are out of P6 scope (need the libtorch tier).

Parity (section 4.1, release-blocking): committed fixture
meridian_input.json -> native Rust reference (tests/meridian_parity.rs,
calls hardware_norm + geometry + rapid_adapt directly) locks
tests/golden/meridian_output.sha256 over the concatenated f32 outputs
(esp32+intel canonical frames, 64-dim geometry vector, rapid-adapt LoRA
weights); pytest (tests/test_meridian.py) runs the same fixture through
the binding and asserts the identical SHA-256. Both pass.

Verified:
  cargo test --features meridian --test meridian_parity -> 2/2 pass
  maturin develop --features meridian + pytest tests/test_meridian.py
    -> 13/13 pass
  default cargo build clean, 0 train/signal/sensing-server refs in the
    default dep graph (gate keeps the base wheel lean).

WHEEL-SIZE FINDING (ADR-185 section 9 / section 1.2): the libtorch risk
the ADR feared is AVOIDED -- wifi-densepose-train's `tch` dep is properly
optional (feature tch-backend, OFF), so no libtorch links. BUT train
still carries NON-optional deps: tokio (rt subset), the five ruvector-*
crates, and wifi-densepose-nn (which itself pulls `ort` / ONNX Runtime +
reqwest/hyper). So a [meridian] wheel exceeds the ADR-117 section 5.4
<=5 MB budget (though lighter than AETHER's axum/tokio server tree). The
clean fix is the same leaf-crate hoist: move the pure inference modules
(geometry, rapid_adapt, eval, hardware_norm) into a tch/tokio/ort-free
leaf crate. A required pre-release follow-up, not a functional blocker;
P2 binds real code and proves parity today.

python/Cargo.toml

@@ -31,6 +31,11 @@ default = []
 # `wifi-densepose-sensing-server` crate (see the honest wheel-size note
 # on that dep below).
 aether = ["dep:wifi-densepose-sensing-server"]
+# ADR-185 P2 — MERIDIAN domain generalization. Binds the tch-free
+# inference/adaptation path only (see the wheel-size note on the deps
+# below). `wifi-densepose-train` is depended on WITHOUT `tch-backend`,
+# so no libtorch is linked.
+meridian = ["dep:wifi-densepose-train", "dep:wifi-densepose-signal"]
 
 [dependencies]
 # PyO3 with abi3-py310 — one compiled binary covers Python 3.10, 3.11,
@@ -75,6 +80,23 @@ numpy = "0.22"
 # hoist is the required pre-release follow-up.
 wifi-densepose-sensing-server = { version = "0.3.0", path = "../v2/crates/wifi-densepose-sensing-server", optional = true, default-features = false }
 
+# ADR-185 P2 — MERIDIAN backing crates (optional, `meridian`-gated).
+#
+# HONEST WHEEL-SIZE NOTE (ADR-185 §9 / §1.2): unlike AETHER, the libtorch
+# risk is AVOIDED here — `wifi-densepose-train`'s `tch` dep is properly
+# optional (feature `tch-backend`, OFF by default), so no libtorch links.
+# BUT `wifi-densepose-train` still carries NON-optional deps: `tokio` (rt
+# subset), the five `ruvector-*` crates, `wifi-densepose-nn`, petgraph,
+# memmap2, indicatif, ndarray-npy, csv, toml, clap. So a `[meridian]`
+# wheel is heavier than the ≤5 MB ADR-117 §5.4 budget (though far lighter
+# than AETHER's axum/tokio server tree). The clean fix is the same
+# leaf-crate hoist: move the pure inference modules (geometry,
+# rapid_adapt, eval, hardware_norm) into a tch/tokio-free leaf crate.
+# `wifi-densepose-signal` is depended on `default-features = false` to
+# drop the optional ndarray-linalg/BLAS chain (Windows-friendly).
+wifi-densepose-train = { version = "0.3.0", path = "../v2/crates/wifi-densepose-train", optional = true, default-features = false }
+wifi-densepose-signal = { version = "0.3.0", path = "../v2/crates/wifi-densepose-signal", optional = true, default-features = false }
+
 [dev-dependencies]
 # ADR-185 §4.1 parity harness — SHA-256 the native-Rust reference
 # embedding and read the committed golden fixture.