Reset time when model is changed

[matbesancon]

I am solving a series of LPs, the difference between them is changes to the objective function only.
The solving time reported by HiGHS seems however not to be reset to 0 and therefore accumulates across iterations

[jajhall]

I'll think about changing this, but it's not an issue

[matbesancon]

We were working on a similar issue in SCIP, the moment where the solving-specific cache is reset is when the model parameters are reset (a coefficient change)

[jajhall]

HiGHS doesn't follow SCIP