…024_1_1_256

Links the btclibs Scrypt PoW TU set (scrypt.cpp + SHA256 family) into
header_chain_test and adds an end-to-end test that runs the real
scrypt_1024_1_1_256 over an 80-byte header, converts the 32-byte LE digest
via u256::from_le_bytes, pins it as a self-derived KAT, and feeds it through
the live full-width satisfaction gate (accept vs pow_limit, reject vs a
tiny target -> proves full-width ordering, not a low64 proxy). Compiles only
the scrypt + SHA256 crypto TUs directly (not the full btclibs archive, whose
siphash TU needs the removed base_uint<256> symbol, and not core/the dgb
OBJECT lib) -- linking an existing scrypt.cpp, no shared source modified.
Same target name, still in both build.yml --target allowlists.

Proves the conversion primitive end-to-end against a genuine Scrypt digest;
does NOT wire scrypt into the production node.cpp ingest path (that lands in
embedded-daemon Phase A, scoped with an exe-smoke).

(cherry picked from commit f49a89e)

The ingest path no longer re-derives the required next-work and rejects on
an nBits-exact mismatch. V36's defining constraint is p2pool-merged-v36
compatibility, and that reference never recomputes parent difficulty -- it
trusts the parent header's declared nBits and checks PoW against it. The
two daemon-independent CheckProofOfWork halves (pow_limit floor +
scrypt(header) <= target) remain and ARE the correct, complete V36
parent-difficulty validation.

Re-derivation is also structurally impossible in a Scrypt-only walk:
DigiByte's live retarget is MultiShield V4, whose averaging window is
GLOBAL across all 5 algos with per-algo adjust + MedianTimePast deltas + /4
damping -- reconstructing it needs full multi-algo header tracking == V37
(5-algo validation). digishield_next_target()/next_retarget_window() are
retained as test scaffolding and a reference for the V37 embedded-daemon
port; the ingest path deliberately no longer calls them.

Test IngestGateEnforcesDigiShieldNextTarget -> IngestRetargetGateDemotedTo-
NoOpForV36: a target != the single-algo DigiShield reference is now ACCEPTED;
the reference value stays computable as scaffolding; PoW + ceiling halves
still reject. header_chain_test 21/21 PASS.

(cherry picked from commit 62398af)

Add the daemon-independent, Scrypt-only-walk-safe half of header time
validation: a Scrypt header's nTime must be STRICTLY GREATER than the median
timestamp of the tip and its (up to) 10 nearest ancestors (DigiByte Core
ContextualCheckBlockHeader "time-too-old"). This is the monotonicity floor
the demoted nBits re-derivation gate used to imply, recovered without any
difficulty recompute -- it reads only timestamps already in the local header
chain, so it is safe on a Scrypt-only walk and needs no V4-MultiShield
multi-algo window.

median_time_past() walks ALL appended headers regardless of algo, mirroring
DGB Core's GetMedianTimePast over the block index: continuity (non-Scrypt)
headers DO contribute their timestamps to the median even though their PoW is
never validated and they are never rejected on the gate (accept-by-continuity
is preserved). Genesis (empty chain) returns INT64_MIN -> unconstrained. A
rejection never mutates the chain (checked before push_back).

+4 tests INTO header_chain_test (no new gtest target -> no build.yml
allowlist change, #143 trap avoided): genesis-unconstrained, strict-greater
rejection at/below median, continuity timestamps feed the median, continuity
header itself bypasses the gate. header_chain_test 25/25 PASS, build EXIT=0.

(cherry picked from commit b084acd)