all: print unsigned values with unsigned conversions

[xanimo]

cppcheck reports eighteen invalidPrintfArgType_sint findings — %d used with unsigned arguments across block.c, headersdb_file.c, spv.c, validation.c, wallet.c, such.c, spvnode.c and tool.c.

None of them misbehave for the values these actually carry. A height or a merkle index would have to exceed INT_MAX to print wrong. So this is correctness, not a bug fix, and I'd rather say that up front than dress it up.

It's worth doing anyway for two reasons.

It's most of what stands between the cppcheck gate and the level its own header says it should reach:

Gating widens to warning once the initial backlog is dispositioned in cppcheck-suppressions.txt

And a suppression would have been the wrong tool. The suppressions file is for false positives with a written reason. These are real findings — just harmless ones. Suppressing them is how a gate quietly stops meaning anything.

Choices

uint32_t fields use PRIu32 rather than %u. The tree already uses the PRI macros in koinu.c, net.c and wallet.c, and it states the width at the call site instead of relying on unsigned int and uint32_t happening to agree. Plain unsigned int loop counters use %u. inttypes.h added where PRIu32 is now used and wasn't already included.

dogecoin_bool arguments were deliberately left alone — it's a uint8_t, which promotes to int, so %d is already correct there, and cppcheck doesn't flag them. Changing those would have been noise.

Where this leaves the backlog

Local cppcheck: 34 warnings → 16. The remainder is already covered by open work:

Remaining	Covered by
4 nullPointerRedundantCheck + 2 ctunullpointer	#404
2 negativeIndex (qr.c), 1 invalidscanf (bip39)	#401
2 uninitvar (OP-TEE false positives)	#403 suppressions
4 internalAstError	cppcheck 2.7 parser limits — absent on the 2.13 the CI image ships

So with #401, #403, #404 and this, the phase-1 backlog is essentially clear and the gate can widen to warning as intended.

81/81.