E2E file read changes

[sustrev]

I encountered a few new E2E files that were more thoroughly anonymized by Heidelberg's software and made a few changes so that these files could be read through. This includes adding a try/except for e2e_binary.chunk_structure.parse(raw) and better handling of now potentially missing patient metadata.

For context, the errors that were observed were mostly

Error in path (parsing) -> magic3
stream read less than specified amount, expected 12, found 0

with raw being b''(which could be handled more specifically by looking at len(raw) before parsing, if that's preferred)
and a few instances of
'ascii' codec can't decode byte 0xc7 in position 0: ordinal not in range(128) with raw as b'\xc7\xd7\xbe\x00\x00\x00\x003\xc4\x00\x00\x00\x00\xbf\xbd\xad\xc1\x00\x00}\xc4\x00\x00\x00\x00\x00\x00\x9f\xc5\x8d\xc6\xac\xc1\x00\x00%\xc4\xba\xc0\r\xc3\x00\x00\x00\x00\x00\x00\xd7\xc4\x83\xc1\x00\x00\x06\xb1?\xc3)\xc4\x00' (again, might be a more graceful way of handling this, but for now skipping is simpler and from the files I've tested on still output as expected).

(I'd been hoping to get my hands on some of these new-anonymization files!)

[marksgraham]

Thanks for this!