mrirecon/bart: version 1.0.00

Changes:
    - doc: Add definition files for singularity containers with BART
    - doc: Add documentation of performance options
    - doc: Add documentation for singularity container
    - doc: Add documentation for environment variables
    - affinereg (new): affine registration
    - bloch (new): Bloch simulation (work-in-progress)
    - coils (new): simulate coil sensitivity maps
    - compress (new): compress data using binary mask
    - cunet (new): conditional U-Net
    - denoise (new): denoising tool
    - estmotions (new): non-rigid registration
    - estscaling (new): estimate a scaling factor from k-space center
    - gmm (new): Gaussian-Mixture models
    - grid (new): compute Cartesian grids
    - grog (new): GROG calibration and gridding (currently only radials)
    - hist (new): compute histograms
    - interpolate (new): interpolate and image after transformation
    - mobasig (sig): forward calculation of physical signal models
    - multicfl (new): combine/split multi-cfl files
    - nlinvnet (new): NLINV-Net reconstruction and training
    - ncalib (new): sensitivity calibration using NLINV
    - phasepole (new): tool to detect phase poles
    - pulse (new): compute radio-frequency pulse shapes
    - raga (new): tool to compute RAGA indices
    - rovir (new): coil compression with region optimization
    - sample (new): sample generative score-based models
    - seq (new): compute MRI sequences
    - stl (new): read/write STL files
    - unwrap (new): simply unwrapping tool
    - tee (new): copy stdin to file and stdout
    - toraw (new): create raw data files
    - trajcor (new): retrospectively correct trajectory based on gradient delays
    - mat2cfl: removed.
    - bin: zero-fill option
    - bin: stream support
    - calc: new function: round
    - ccapply: realtime geometric coil compression
    - conway: stream support
    - copy: stream support
    - copy: support for reordering of dimensions
    - epg: make refocusing FA accessible from CLI
    - ecalib: use Nyström approximation
    - ecalib: economic mode for 3D coils
    - fovshift: pass shift as file
    - invert: option for regularization
    - join: stream support
    - mandelbrot: stream support
    - moba: IR-MECO reconstruction
    - moba: finalize transparent gpu support in moba
    - moba: change scaling and command line for T2 moba to be consistent with mobafit
    - mobafit: CEST fitting
    - mobafit: simulation based fitting (Bloch + Bloch-McConnell)
    - nnet: support for nnU-Net for segmenation
    - nlinv: basis support, temporal regularization, builtin nufft
    - nlinv: stream support
    - nufft: fix compatibility to v0.8.00
    - pics: use of prox in last iteration of FISTA
    - pics: (TE-)ASL reconstruction
    - pics: Euler-Maruyama
    - reconet: fix compatibility to v0.8.00
    - show: support for printing real part and index
    - signal: enable the input of fat fraction
    - signal: generating function-based FSE model
    - signal: buxton model
    - signal: option to toggle short TR approximation
    - sim: CEST sequence
    - traj: option to create RAGA index file
    - twixread: optional read PMU signal
    - twixread: read noise scans
    - rtreco.sh: new script for realtime reconstruction
    - library: remove divergence free wavelets
    - library (num): stable and improved random number generation with Philox-4x32
    - library (num): GPU support for gaussian random numbers
    - library (num): rename md_(de)compress -> md_mask_(de)compress
    - library (num): new functions: md_(z)round(2) functions, md_zsetnanzero(2)
    - library (num): MPI support for md_copy with strides, and md_zadd2
    - library (num): new LAPACK wrappers (svd, eigenvalues, solvers)
    - library (num): matrix logarithm of upper triangular matrix
    - library (num): matrix square root of triangular matrix
    - library (num): complex support for matrix exponentials
    - library (num): new special functions: Legendre polynomials
    - library (num): new special functions: simplified hyp2f1 function
    - library (num): new special functions: approximation of gamma function
    - library (num): Gauss-Legendre quadrature
    - library (num): Crank Nicolson
    - library (num): functions for Gaussian mixture models
    - library (ismrmd): new functions to parse ismrmrd streams
    - library (iter): Euler Mayurama
    - library (linop): shift, slice, hankelization, reshape with flags
    - library (nn): cast linop to nn
    - library (nn): support for leaky ReLU
    - library (nn): normalization layer
    - library (nn): normalized loss
    - library (nlops): checkpointing container
    - general: move to codeberg
    - general: remove TOOLBOX_PATH and DEBUG_LEVEL (use the new ones with BART_ prefix)
    - general: stream protocol implementation + tests
    - general: experimental support for delayed computation
    - general: Windows DLLs