Vanadis - Nested loop for resetting temp registers

[Anunalla]

In Vanadis Tick() function, the temp registers are being reset N^2 times for N threads.
Here, the register reset function is being called within a loop to go over all threads,

sst-elements/src/sst/elements/vanadis/vanadis.cc

Line 1472 in fd22b77

resetRegisterUseTemps(thread_decoders[i]->countISAIntReg(), thread_decoders[i]->countISAFPReg());

however, the reset function itself hosts a loop to go over all the threads.

sst-elements/src/sst/elements/vanadis/vanadis.cc

Line 507 in fd22b77

VANADIS_COMPONENT::resetRegisterUseTemps(const uint16_t int_reg_count, const uint16_t fp_reg_count)

Profiling the simulation shows that the simulator spends 10% of its time on this function.

Details on profiling experiment:
System: Ubuntu 20.04
SST: Version 14.1
AARCH: riscv64
Cross compiler: MUSL 10 riscv64-linux-musl (https://more.musl.cc/10/x86_64-linux-musl/)
Test: Matrix multiplication using pthreads (Number of worker threads-2, Number of matrix multiplications - 2, Matrix size=64x64, )
Config file: basic_vanadis.py
Environment variables:

VANADIS_VERBOSE=0
VANADIS_OS_VERBOSE=0
VANADIS_NUM_HW_THREADS=16
VANADIS_EXE= /path/to/my/custom/test