reject short sequence in cms content-type parsers

[rootvector2]

the cms content-type decoders SignedData, EnvelopedData, AuthenticatedData, AuthEnvelopedData and EncryptedData read their mandatory fields with no lower-bound size check, so a ContentInfo whose inner content is an empty or too-short SEQUENCE leaks a NoSuchElementException/ArrayIndexOutOfBoundsException out of the throws CMSException contract of CMSSignedData(byte[]) and its siblings (the IllegalArgumentException CMSEncryptedData documents) — found auditing the CMS parse entry points; each now rejects a short sequence up front like CompressedData/DigestedData already do, covering the interspersed case where a leading OPTIONAL is claimed and the mandatory fields are then truncated.

AI tooling was used to help prepare this change.