Fix a first-check-then-do race condition during framework shutdown.

[PengZheng]

static void *fw_eventDispatcher(void *fw) {
    framework_pt framework = (framework_pt) fw;

    celixThreadMutex_lock(&framework->dispatcher.mutex);
    bool active = framework->dispatcher.active;
    celixThreadMutex_unlock(&framework->dispatcher.mutex);

    while (active) { //first check active then handle events
        fw_handleEvents(framework);
        celixThreadMutex_lock(&framework->dispatcher.mutex);
        active = framework->dispatcher.active;
        celixThreadMutex_unlock(&framework->dispatcher.mutex);
    }
    //omitted
}

If shell_cmd happens to shutdown the framework during this first-check-then-do race window, unnecessary 1 second delay follows.

[codecov-commenter]

Codecov Report

Merging #402 (aa98685) into master (3b651da) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #402      +/-   ##
==========================================
- Coverage   73.12%   73.12%   -0.01%     
==========================================
  Files         205      205              
  Lines       31272    31272              
==========================================
- Hits        22869    22867       -2     
- Misses       8403     8405       +2     

Impacted Files	Coverage Δ
libs/framework/src/framework.c	79.54% <100.00%> (ø)
libs/utils/src/hash_map.c	93.53% <0.00%> (-0.57%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3b651da...aa98685. Read the comment docs.

[PengZheng]

About the shell test deadlock: https://github.com/apache/celix/runs/5562712613?check_suite_focus=true#step:5:3036
I have not been able to reproduce it in my local environment.

[pnoltes]

About the shell test deadlock: https://github.com/apache/celix/runs/5562712613?check_suite_focus=true#step:5:3036
I have not been able to reproduce it in my local environment.

I also have not been able to reproduce this locally.