Improve iClass SIO and legacy credential detection reliability

[nvx]

Now relies on the legacy config block for SR detection and the end-of-SIO detection no longer partially cuts off the SIO for any dumps I have.

The previous code would cut off some SIOs mid way through if they happened to contain the wrong bytes mid way through.

Also fixes a couple of super minor display quirks I noticed while running hf iclass view on all my dumps - for example if the stars aligned on a SE dump it could in some cases try and decode block 7 as if it were a legacy dump resulting in rather confusing output, and a case where hf iclass decrypt would attempt to use a card helper even if the flag wasn't set (again if the stars aligned with the right block data).

[iceman1001]

Awedsome, there were some interesting fixes, I merge as is.

Feedback,

• The reverse strstr got me confused. Maybe add a comment about the two functions now in util.c.
• the SR BIT 0xA0 Should be a MACRO define, to denote what it is.
• help text for default start 3 and 6, should be in the general description. Otherwise the column text becomes too large (save screen space)

I see you must have had some unsecure tags to play with in order to trigger them :)