Ledger Bitcoin app

Fix missing return value checks for point_add calls

Also made the cast from size_t to int explicit, and fixed PRINTF
message that had an off-by-one.

Merge pull request #539 from LedgerHQ/fix-stack-exhaustion

Fix stack exhaustion

Add functional tests for the policy depth limits

Limit the amount of nesting for `thresh` fragments to 4

compute_thresh_ops and compute_thresh_stacksize each need two arrays of
MAX_N_IN_THRESH + 2 counters. Since they were inlined by the compiler,
they bloat the size of the stack frame of
compute_miniscript_policy_ext_info, which vastly reduces the stack usage
on usual policies.

Yet, stack usage would remain very large on policies that recursively
nest 'thresh' fragments.
Therefore, we add a limit of 4 nested thresh expressions, by keeping
track in the parsing context.

Move the wallet confirmation stage out of the handler's frame

By not inlining the called functions, the memory occupation can be
substantially improved, as a lot of the data used during validation
are no longer needed during rendering and UI.

Count miniscript wrappers in the parser's recursion depth limit

parse_script bounds the depth of the parsed policy with
MAX_PARSE_SCRIPT_RECURSION_DEPTH, but the miniscript wrappers were
not charged to that budget; this can cause stack exhaustion in other
functions that process the parsed AST recursively.

Lowers MAX_N_IN_THRESH to 24

It is extremely unlikely to be hit in practice, and has a large memory
impact because of the dynamic programming tables of compute_thresh_ops()
and compute_thresh_stacksize()

The dynamic programming tables of compute_thresh_ops() and
compute_thresh_stacksize() are sized for MAX_N_IN_THRESH, causing them
to contribute a substantial stack usage, especially with nested
thresh fragments.

Fix wrong comment in policy_node_thresh_t

Disable musig in key expressions of musig_a/sortedmulti_a

The parsing of such fragments is not correctly handled in the
current state. Therefore, we cleanly disable it until the parser
is generalized.

Harden prevout tx parsing checks

Sign consistency for parser steps; removed stale comment

Fix taptree hash computation in test_utils

Bump version to 2.5.0

Update snapshots

For external input amounts, only show if trustworthy

Because legacy and SegwitV0 sighash algorithm doesn't commit to the
other inputs' amounts, the amount of external inputs is only
trustworthy if validated via the non-witness-utxo.
That is not a problem for taproot.

Update snapshots

Tests for showing transactions with external inputs

[auto-update] Update Ragger snapshots

[auto-update] Update Ragger snapshots

Using also_delete_snapshots parameter

Adding codecov configuration file

Update test snapshots

Add testcase for registering a policy with too high confusion score

The test verifies that register_wallet still succeeds, but showing
the raw descriptor template (fallback to usual behavior) rather
than the cleartext version.

Also, slightly refactored the shared code across unit tests.

Show cleartext representation of wallet policy, when possible

Wires the cleartext engine into the register-wallet flow: the handler
computes the confusion score and, when below the threshold, renders the
cleartext spending-path lines shown before the raw descriptor template.

For multisig wallet policies specifically, this also simplifies the UX by
omitting the raw descriptor template altogether. This is safe for such
simple policies, with very little ambiguity. In the long term, we might be
able to extend this to more complex wallet policies, but that needs to be
done intentionally, and possibly be an opt-in feature that is explicitly
requested by software wallets that adapt their own UX accordingly.

Unit tests of register_wallet had to be adapted slightly to take into
account that the descriptor is not always shown.

Add testcases with snapshots for multisig k-of-n and several taproot leaves