mhi: core: avoid triggering SYSERR if device is already in RDDM

If device gets a fatal error and a kernel panic follows, MHI host
would trigger the device to RDDM when it has already entered
RDDM. This would result in a stall on the device blocking ramdump
collection. Avoid sending SYS_ERROR to trigger device to RDDM
from kernel panic path if it is already in RDDM.

CRs-Fixed: 2425417
Change-Id: I86132e2d485f57c3f68dc97faeb1c0f90bde8832
Acked-by: Bhaumik Vasav Bhatt <bbhatt@qti.qualcomm.com>
Signed-off-by: Sujeev Dias <sdias@codeaurora.org>