Proposal: Add automatic bootstrap download on first run for raptoreumd

[Germardies]

Automatic Bootstrap Download on First Run for raptoreumd (no QT)

This PR adds automatic blockchain bootstrap downloading when Raptoreum Core is started for the first time with an empty data directory.

What this does

• Detects first run by checking if blocks/blk00000.dat is missing
• Automatically downloads bootstrap.zip from bootstrap.raptoreum.com
• Extracts the archive into the data directory
• Downloads powcache.dat for faster initial

This is purely a suggestion for improvement!

[JSanchezFDZ]

Reviewed the auto-bootstrap downloader. The intent is useful, but there are a few
blocking concerns, mostly around security:

1. Duplicate / mixed scope. This PR's head (feature/auto-bootstrap,
   eb4a1ea) is the exact same commit as #437, and it bundles two unrelated
   features — the asset feeAddress work in rpc/rpcassets.cpp and the bootstrap
   downloader — plus incidental changes (lockedpool.cpp, bip32.h, .gitignore).
   Please split into one focused PR per feature and close the duplicate; as-is
   neither change can be reviewed or merged cleanly.

2. No integrity/authenticity check (security). DownloadFile() fetches
   bootstrap.zip / powcache.dat over HTTPS but nothing verifies the payload
   (no SHA-256, no signature). A compromised or MITM'd bootstrap host could feed a
   crafted snapshot. Please verify a published, ideally signed, SHA-256 of the
   archive before extracting and abort on mismatch.

3. Path traversal / zip-slip (security). ExtractArchive() uses
   archive_write_disk_new() without archive_write_disk_set_options(...). Set
   ARCHIVE_EXTRACT_SECURE_NODOTDOT | ARCHIVE_EXTRACT_SECURE_SYMLINKS | ARCHIVE_EXTRACT_SECURE_NOABSOLUTEPATHS; otherwise an archive entry with ../,
   an absolute path, or a symlink can write outside the data directory. The manual
   destDir + "/" + archive_entry_pathname() concatenation does not prevent this.

4. Error handling. Return values of archive_write_header() /
   archive_write_data_block() are ignored, and ExtractArchive() returns true
   even when the read loop aborts on error, so it can report "Extraction complete"
   after a partial/failed extract. DownloadFile() also doesn't set
   CURLOPT_FAILONERROR, so an HTTP 4xx/5xx error page is written into
   bootstrap.zip and only fails later at extraction. Consider CURLOPT_FAILONERROR
   and explicit CURLOPT_SSL_VERIFYPEER/CURLOPT_SSL_VERIFYHOST.

5. New daemon dependencies. This adds -lcurl -larchive to raptoreumd,
   which is a meaningful new attack/build surface and complicates depends and
   reproducible builds (related to #151). Worth a maintainer decision before
   pulling libcurl into the daemon.

6. Opt-in/opt-out. It triggers automatically on first run (detecting a missing
   blocks/blk00000.dat). Please make it opt-in or at least disableable via a flag,
   and note that -reindex, pruned, or custom datadir layouts may misdetect a
   "first run".