Memory-safety fixes: uninitialized reads and a realloc leak

[matejk]

Two commits:

• Reads of uninitialized values on error paths (4 sites): tape_get_pews leaves *pews unset on -LTFS_UNSUPPORTED while the caller computes pews + 10; ltfs_profiler_set tested an unset ret when the volume had neither an iosched nor a device handle; _get_dump derived a transfer length from cap_buf without checking the READ BUFFER result; the ICU normalize callers compared the output pointer against the input before checking the return code, reading uninitialized memory and potentially leaking the input buffer.
• Key-list leak in the simple KMI: realloc was assigned straight back to priv.dk_list, so a failure overwrote the only pointer to the existing buffer with NULL and leaked it; the original pointer is now kept and freed on the error path.

[XV02]

Thanks for all your work, just some correctness comments :DD

[matejk]

Thanks — addressed:

• The comments on the initializations and NULL checks are removed.
• simple.c: back to the single error path; the original pointer is kept in previous and freed there. free(NULL) is a no-op on the calloc branch, so no extra condition is needed.

make check 14/14.

[XV02]

Thanks for addressing all my comments, LGTM! :DD