- The -focus option was not labelled as type_sequence_float(), leading to
  a warning if the first value is negative, e.g.:

      mrview: [WARNING] Value "-1,2,3" is being used as the expected argument
      for option "-focus"; is this what you intended?

- It should be allowed to use the -target option multiple times.

This simply clamps values of the input data at zero before taking their
mean per-shell, which seems to fix issues raised in #2248.

Discussed in #2248

…_data

dwi2response dhollander: fix handling of data with negative values

Rebase to latest master

Use erase-remove idiom, otherwise particles will be duplicated instead of removed.

When tckglobal command was run with -niter 10B the numbers of iterations,
that run were maximally 1410065408 instead of 10000000000.
This change fixes this parsing issue.

…moval

tckglobal: Fix bug that caused particles in a voxel to be duplicated when tried to be removed

tckglobal: Fix erroneous -niter parsing for values bigger that 1.4e9

…astic case

Inference: Fix for exotic usage

…s_target

mrview: fix handling of -focus and -target options

…tics_parsing

python: fix handling of N/A output in image.statistics()

Downstream of #1800, which introduced the functionality of determining the optimal reordering of directions for all possible selections of first volume, and choosing the optimal reordering from that set. Algorithm produces a segmentation fault if fewer than six input directions. This change forces the first volume in the input to be the first volume in the output in such scenarios (as there is no objective criterion by which to determine which direction is best to use first in this scenario).

Comparisons between signed and unsigned integers introduced in 035a98f.

dirorder: Fix segfault

…_line_option

add two new command lines option in mrview for tractogram

For help page interfaces of algorithmm module wrapper scripts, do not insert the prospect of command-line options in between the wrapper name and the algorithm selection, as such usage will not work.
As reported in #2292.

In instances where the input data contain non-finite values (regardless of whether or not such data are excluded from statistical inference via a mask), and at least one F-test is performed, writing the absolute and standardised effect size as NaN (as such cannot be meaningfully calculated for an F-test) was incorrectly indexed, leading to a segmentation fault.

mrview: fix glitch in handling of clip plane selection

Fix help page for algorithm module

GLM: Fix all_stats() segfault

more verbose error messages in population_template

mrregister: fix lmax check for nonlinear multi contrast registration

Conflicts:
	.github/workflows/checks.yml
	lib/mrtrix3/run.py