Add Bounding Box to Field Line Integration

[dpanici]

• Adds a bounding box to the field_line_integrate defined by bounds_R and bounds_Z keyword arguments, which form a hollow cylindrical bounding box. If the field line trajectory exits these bounds, the RHS will be multiplied by exp(-(R**2+Z**2)) to stop the trajectory and prevent tracing the field line out to infinity, which is both costly and unnecessary when making a Poincare plot, which is the principle purpose of the function.

In one example I had when tracing the precise_QA with a coilset that had some diverging field lines, it took 1/3 of the time compared to without using bounds, so seems to work fine. Defaults to no bounds.

@ddudt this PR is not addressing anything that happens when a SplineMagneticField is used and the field lines exit its domain of validity when extrap=False unless you specify the bounding box to lie within its domain of validity, in which case it should stop the integration soon after it exits the bounding box and possibly before it exits the field's interpolation domain. If there are ideas of how to treat that then we can make an issue and a separate PR

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.05%. Comparing base (664a562) to head (7e48dbc).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #908   +/-   ##
=======================================
  Coverage   95.05%   95.05%           
=======================================
  Files          83       83           
  Lines       20966    20970    +4     
=======================================
+ Hits        19929    19934    +5     
+ Misses       1037     1036    -1     

Files	Coverage Δ
desc/magnetic_fields.py	96.97% <100.00%> (+0.01%)	⬆️

... and 1 file with indirect coverage changes

[github-actions]

|             benchmark_name             |         dt(%)          |         dt(s)          |        t_new(s)        |        t_old(s)        | 
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
 test_build_transform_fft_lowres         |     -0.41 +/- 2.09     | -5.21e-05 +/- 2.66e-04 |  1.27e-02 +/- 1.2e-04  |  1.28e-02 +/- 2.4e-04  |
 test_build_transform_fft_midres         |     +1.36 +/- 1.45     | +1.25e-03 +/- 1.34e-03 |  9.38e-02 +/- 8.9e-04  |  9.25e-02 +/- 1.0e-03  |
 test_build_transform_fft_highres        |     +1.34 +/- 0.85     | +6.19e-03 +/- 3.91e-03 |  4.68e-01 +/- 3.3e-03  |  4.62e-01 +/- 2.1e-03  |
 test_equilibrium_init_lowres            |     -0.23 +/- 2.45     | -1.36e-03 +/- 1.48e-02 |  6.03e-01 +/- 5.4e-03  |  6.04e-01 +/- 1.4e-02  |
 test_equilibrium_init_medres            |     +0.10 +/- 2.07     | +1.25e-03 +/- 2.59e-02 |  1.25e+00 +/- 1.5e-02  |  1.25e+00 +/- 2.1e-02  |
 test_equilibrium_init_highres           |     -0.23 +/- 1.02     | -9.66e-03 +/- 4.28e-02 |  4.17e+00 +/- 3.1e-02  |  4.17e+00 +/- 3.0e-02  |
 test_objective_compile_dshape_current   |     -1.00 +/- 2.43     | -4.69e-02 +/- 1.13e-01 |  4.63e+00 +/- 5.7e-02  |  4.68e+00 +/- 9.8e-02  |
 test_objective_compile_atf              |     -0.93 +/- 2.63     | -8.64e-02 +/- 2.45e-01 |  9.21e+00 +/- 1.9e-01  |  9.30e+00 +/- 1.5e-01  |
 test_objective_compute_dshape_current   |     -0.67 +/- 1.71     | -4.38e-05 +/- 1.11e-04 |  6.46e-03 +/- 7.4e-05  |  6.50e-03 +/- 8.4e-05  |
 test_objective_compute_atf              |     +7.63 +/- 5.06     | +1.01e-03 +/- 6.72e-04 |  1.43e-02 +/- 4.9e-04  |  1.33e-02 +/- 4.6e-04  |
 test_objective_jac_dshape_current       |     +5.75 +/- 7.04     | +2.54e-03 +/- 3.11e-03 |  4.68e-02 +/- 1.5e-03  |  4.42e-02 +/- 2.7e-03  |
 test_objective_jac_atf                  |     +1.77 +/- 3.76     | +3.52e-02 +/- 7.47e-02 |  2.02e+00 +/- 2.8e-02  |  1.98e+00 +/- 6.9e-02  |
 test_perturb_1                          |     -2.79 +/- 2.49     | -3.91e-01 +/- 3.48e-01 |  1.36e+01 +/- 3.3e-01  |  1.40e+01 +/- 1.1e-01  |
 test_perturb_2                          |     +1.81 +/- 1.94     | +3.63e-01 +/- 3.89e-01 |  2.05e+01 +/- 3.1e-01  |  2.01e+01 +/- 2.3e-01  |
 test_proximal_jac_atf                   |     -0.09 +/- 1.05     | -5.87e-03 +/- 7.14e-02 |  6.80e+00 +/- 4.8e-02  |  6.81e+00 +/- 5.3e-02  |
 test_proximal_freeb_compute             |     +0.41 +/- 0.67     | +5.08e-04 +/- 8.18e-04 |  1.23e-01 +/- 5.7e-04  |  1.23e-01 +/- 5.8e-04  |
 test_proximal_freeb_jac                 |     +2.50 +/- 4.17     | +2.04e-01 +/- 3.40e-01 |  8.35e+00 +/- 1.9e-01  |  8.15e+00 +/- 2.8e-01  |

[dpanici]

we don't expect the integration to stop immediately after the bound is exited, but it should stop soon after depending on how fast the RHS decays. I made the decay very fast so we can test it, but if for some reason we should lessen the decay rates, then we will need to make these test tols here less tight