MIP Worker Refactor

[Opt-Mucca]

No description provided.

[Opt-Mucca]

@jajhall : Would this be the timer issue you mentioned on Tuesday (when the same highs instance is run multiple times)?

intlinprog(prob);
Running HiGHS 1.14.0: Copyright (c) 2026 under MIT licence terms
MIP has 212 rows; 260 cols; 8619 nonzeros; 259 integer variables (259 binary)
Coefficient ranges:
Matrix [1e+00, 7e+02]
Cost [1e+00, 1e+00]
Bound [1e+00, 1e+00]
RHS [1e+00, 3e+02]
ERROR: Option 'threads' is set to 4 but global scheduler has already been initialized to use 2 threads. The previous scheduler instance can be destroyed by calling Highs::resetGlobalScheduler().
Error using intlinprog (line 108)
INTLINPROG has stopped because it encountered an internal error. We are sorry for the inconvenience.

That's some output from the test of @fwesselm of this branch with mip_search_concurrency > 0

[jajhall]

No, this is due to someone running a second Highs instance with threads set to a different value

[Opt-Mucca]

@fwesselm I've updated the branch so the interrupt callback is not flagged when parallelLockActive, and I've made most of HighsMipWorker private (a point you raised in an earlier comment).
I believe this is on track to be merged next week (after I change mip_search_concurrency=0 to default)