Incorrect crest conformer search energy when using mtd and optimization method

[jevandezande]

Following the custom conformational search example, the dynamics method affects the final energy after optimization.

To Reproduce geom.xyz

5
Methane
C     0.0000000000    0.0000000000    0.0000000000
H     0.6250000000    0.6250000000    0.6250000000
H    -0.6250000000   -0.6250000000    0.6250000000
H    -0.6250000000    0.6250000000   -0.6250000000
H     0.6250000000   -0.6250000000   -0.6250000000

input.toml

input="geom.xyz"
runtype="imtd-gc"
multilevelopt = false

[calculation]
[[calculation.level]]
# MTD
method="gfnff"
weight=1.0

[[calculation.level]]
# Opt
method="gfn2"

[dynamics]
active=[1]

MTD Method	Energy
GFN-FF	-4.80606
GFN0	-8.53465
GFN1	-8.44947
GFN2	-8.35044

[pprcht]

Thanks for noticing this!