trying to add an option -tstrajectory_max_cps

refactor the code a little bit

fix the memory access bug when creating revolve object

cleanup trailing whitespaces

forgot to add revovle source files

results verified to be correct

allow recomputing the solution when not all time steps are checkpointed in memory
for testing, the checkpoints are randomly determined

clean up spaces after comma

Use PetscFunctionBegin instead of PetscFunctionBeginUser

fix a bug related to stack size; allow overwriting top elements in stack

bug fix for checking stack overflow

add in-memory checkpointing, and it works :)

correct more names for TSTrajectory and add TS to the argument list when creating TSTrajectory

use tj to name a TSTrajectory variable

remove debug code

can run with multiple mpi processes

fix memory issues

add one call to adjoint monitor in the end of TSAdjointSolve to visualize the sensitivity result at the starting time

TSTrajectory singlefile does not save the stages solutions only the solution at each timestep

trajectory now saves current time

PetscHeaderCreate: Shorten argument list and optimize memory allocation

* Remove type arguments from the macro
* Use a single memory allocation call instead of three

updated PETSc directory layout to match standard packaging strategies

include/petsc finclude,private,mpiuni
lib/petsc conf
bin/petsc*

Viewer: Now ViewerRead() takes an extra argument to return the number of items read
- An error is thrown for insufficient data unless the extra argument is non-NULL

Viewer: PetscViewerRead() should return the number of items actually read
- This also applies to all its implementations