fix: handle zero-temperature rescaling

[galjos]

Closes #345.

• keep zero-to-zero velocity rescaling finite
• reject positive targets from zero kinetic temperature
• apply the same rule to hard reset, Berendsen, and velocity-rescaling paths

Tests:

• testResetKinetics
• testThermostat

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 89.84%. Comparing base (109a355) to head (cc9ed25).

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #350      +/-   ##
==========================================
+ Coverage   89.83%   89.84%   +0.01%     
==========================================
  Files         291      291              
  Lines       11535    11551      +16     
  Branches     3601     3609       +8     
==========================================
+ Hits        10362    10378      +16     
  Misses       1131     1131              
  Partials       42       42              

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[pq-perf-bot]

⚡ Performance (instruction count) — ✅ no regressions

per-benchmark breakdown

benchmark	base Ir	PR Ir	Δ
bondedForces	41.47M	41.47M	-0.00%	✅
boxTransforms	10.67M	10.67M	+0.00%	✅
constraints	11.41M	11.41M	+0.00%	✅
coulombKernel	5.80M	5.80M	+0.00%	✅
forceKernel	15.35M	15.35M	+0.00%	✅
integrator	32.19M	32.19M	+0.00%	✅
kinetics	9.53M	9.53M	+0.00%	✅
linearAlgebra	2.07M	2.07M	+0.00%	✅
nonCoulombPairs	5.48M	5.48M	+0.00%	✅
shiftVector	5.81M	5.81M	+0.00%	✅
virial	11.64M	11.64M	-0.07%	✅

_{Deterministic callgrind instruction counts vs the base branch; gated at ±2%. Not wall-clock.}

[ape33]

what about updating this second if-statement condition to

if (isZero(_temperature) && !isZero(targetTemp))

then you could omit the first if all together

[97gamjak]

see ape comment