host1x_emu: DT Property based syncpoint memory

This change uses DT property to select syncpoint memory.
If property "syncpoints-mem" is defined hypervisor syncpoint
memory mechanism is used. Otherwise memory is allocated from
guest kernel.

Previously this was based on code level MACRO. Changing it to
DT simplify selection of syncpoint memory.

Jira HOSTX-5982

Change-Id: I5e26ef2d42d5d01d6aff0f2b43fd11c23218057d
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3329245
Reviewed-by: Leslin Varghese <lvarghese@nvidia.com>
Reviewed-by: Sourab Gupta <sourabg@nvidia.com>
Tested-by: Amitabh Dutta <amitabhd@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Amitabh Dutta <amitabhd@nvidia.com>

drivers/gpu/host1x-emu/dev.h

@@ -60,6 +60,7 @@ struct host1x {
 #ifdef HOST1X_EMU_HRTIMER_FENCE_SCAN
 	unsigned int hr_polling_intrval;
 #endif
+    bool hv_syncpt_mem;
 #ifdef HOST1X_EMU_HYPERVISOR
     void __iomem *syncpt_va_apt;   /* syncpoint apperture mapped in kernel space */
 #else