Fix drug recommandation drug code and padding.

[Logiquo]

Fix #1137

[jhnwu3]

tldr; Would be good to just delete the .codex file.

I'm curious about the logic wit the test casing with the FakeLegacyPatient class here, but not too concerned at the moment.

[jhnwu3]

I forgot to mention, sequence processor can also handle the code mappings now too. Don't think it matters that much here, but worth sharing.

class SequenceProcessor(FeatureProcessor, TokenProcessorInterface):
    """Feature processor for encoding categorical sequences.

    Encodes medical codes (e.g., diagnoses, procedures) into numerical
    indices. Supports single or multiple tokens and can build vocabulary
    on the fly if not provided.

    Args:
        code_mapping: optional tuple of (source_vocabulary, target_vocabulary)
            to map raw codes to a grouped vocabulary before tokenizing.
            Uses ``pyhealth.medcode.CrossMap`` internally. For example,
            ``("ICD9CM", "CCSCM")`` maps ~128K ICD-9 diagnosis codes to
            ~280 CCS categories, and ``("NDC", "ATC")`` maps ~940K drug
            codes to ~5K ATC categories. When None (default), codes are
            used as-is with no change to existing behavior.

    Examples:
        >>> proc = SequenceProcessor()  # no mapping, same as before
        >>> proc = SequenceProcessor(code_mapping=("ICD9CM", "CCSCM"))
    """

[Logiquo]

This is a bit tricky because NestedSequence & MultiLabel does not yet support this. May be better to left this into a future PR.

[jhnwu3]

new changes lgtm